South Australian election wrap-up

The South Australian election guide has now been put to bed with each electorate entry appended with a detailed post-match summary. For the first time, I have taken the effort to carefully scrutinise booth results in each seat. Heavy-duty hair-splitters might care to note that I was compelled to base these comparisons on incomplete election night figures from 2002 as the State Electorate Office did not update the two-party preferred booth results on its website beyond that point. They could only be obtained from the Statistical Returns document which, being a PDF file, could not readily be pasted into Excel. As a result, booth comparisons might be out by 1 per cent or so. As you’re all no doubt aware, the last two seats that were in doubt resulted in the re-election of the Liberals’ Graham Gunn in Stuart and Labor-turned-Greens-turned-independent MP Kris Hanna in Mitchell. To the best of my knowledge, the only observer who publicly predicted the latter outcome was Poll Bludger commenter Dave S.

The upper house has passed without comment on this site since the days immediately after the election, but the final result was as predicted then – four Labor, three Liberal, two No Pokies, one Family First and one Greens. The SEO doesn’t have a breakdown of the preference distribution, but we can always plug the results into the election calculator at Upperhouse.Info, bearing in mind that it irons out the complication of below-the-line votes. All but the Family First and Greens candidates were elected off the primary vote, after which Family First had 0.60 of a quota, the Greens 0.51, the third No Pokies candidate (who reportedly promised his wife he had no chance of winning) 0.46, the fifth Labor candidate 0.31 and the Australian Democrats 0.21. Family First were boosted to a quota after distribution of preferences from minor candidates and the fourth Liberal. That left the Greens on 0.56, the Democrats and No Pokies on 0.48 and Labor on 0.41. If the Democrats had been able to persuade Labor to put them ahead of No Pokies, Labor’s subsequent elimination would have put Kate Reynolds ahead of the Greens and subsequently back into parliament with No Pokies preferences. Instead, Labor’s elimination left the Democrats well to the rear of No Pokies and their own preferences delivered victory to Mark Parnell, the first South Australian Greens candidate ever to win a seat at an election.

One day, I will get around to producing retrospective guides to the South Australian upper house and the Tasmanian election. Speaking of Tasmania, the next item of business is the periodical Legislative Council elections for the seats of Rowallan (a lay-down misere for independent incumbent Greg Hall) and Wellington (a taller order for Labor member Doug Parkinson), which will be held on May 6.

UPDATE: If you’re looking for more of a big-picture view of the South Australian election, you could do a lot worse than this paper by Geoff Anderson and Haydon Manning of Flinders University for the Australian National University’s Democratic Audit. Of particular interest is the graph on page three indicating a clear long-term trend of increasing minor party voting in the upper house. Upperhouse.Info sheds more light on this with a table outlining the "desertion rates" of parties’ supporters switching their vote in the upper house. In view of all this, it is clear why Mike Rann would like to abolish the chamber through a referendum he plans to hold in conjunction with the next election. It is equally clear that while The Advertiser might be dopey enough to support abolition, the public will not be.

Author: William Bowe

William Bowe is a Perth-based election analyst and occasional teacher of political science. His blog, The Poll Bludger, has existed in one form or another since 2004, and is one of the most heavily trafficked websites on Australian politics.

13 comments on “South Australian election wrap-up”

  1. William,

    How does one gat “the Statistical Returns document”. It is not yet posted on the seo site. If you could forward it that would be appreciated.

  2. Ray,

    I dont think William has 2006’s statistical returns document, only the 2002 one (which is on the SEO site). Ofcourse if im wrong, and william does have it, id also like a copy (got a few blanks that need filling in my spreadsheet).

    William, regarding your assessment of the hartley result “The 6.7 per cent swing in Hartley was at the lower end of the Adelaide scale, but it was more than enough to make the difference in this finely poised marginal seat. The two-party swing was slightly higher than might have been expected from the primary vote figures, which had Labor up 6.0 per cent to 45.0 per cent and Liberal down 4.1 per cent to 39.5 per cent, due to a reversal of the donkey vote from 2002, a higher vote for the Greens (from 4.2 per cent to 6.7 per cent) and a lower vote for Family First (from 4.6 per cent to 4.2 per cent – one of only two electorates where their vote fell). Booth results were variable and did not follow a pattern consistent with the aforementioned north-south divide. In the north – supposedly the stronghold of the Italian community to which Grace Portolesi directed so much effort – there was a 6.9 per cent swing in Hectorville but no swing at all in Glynde; in the more affluent south, Kensington Gardens swung only 3.3 per cent while Magill shifted a solid 8.8 per cent.”

    regarding the fluctuations in votes, they are all mainly the result of redistribtions. The two northern booths (felixstow and glynde) which produced smaller swings Id say would of been the result of the addition to the electorate of the small section in payneham which is basically one big retirement village. (mind you i cant be certain about that assessment, as both booth produced lower turnouts than 02 – possibly the most puzzling result, as 150 voters who voted at marden west also appeared to have dissapeared meaning the North west booths recorded a drop of 350 voters despite having a redistribution moving several hundred into the electorate)

    As for the southern booths and the difference between Magill and the two kensington booths, firstly there is a big difference between south and north of magill road in ‘poshness’, the two kensington booths cant really be compared with the magill one. The reason that the magill swing was so large was because of the redistribution. A large section in magill was moved (this booth recorded a raise of about 500 votes this time round) into the electorate and thus the incumbant factor was maginalised.

  3. Ray: like Cameron says, I was referring to the 2002 document. The 2006 one won’t be available for a few months.

    Cameron: thanks very much for all that, I have substantially revised the entry on the basis of this information. The upshot of it all is that the swing was concentrated in the north and the southern booths indeed failed to shift much, which is consistent with what I said in my last “campaign update”.

    If anyone else out there can offer this sort of detail on local results, their thoughts would be greatly appreciated.

  4. Just on the addition of a retirement village, as most elderly have a booth bought to them, they are counted as postal votes

  5. I thought that was only for those that are living in assisted living facilities (ie where they need help for day to day things). The majority of the added section was just a retirement villiage where the residents live in self sustained units. While im sure many do vote by postal vote so they dont have to vote on polling day, i didn’t think they had special teams visit them.

  6. it depends on the retirement centre. My grandparents lived in self sustained units but still had a team out cos the village asked for one.

  7. I felt I should make an apperance on this site and thank Pollbludger for recognising my correct prediction of Kris Hanna winning as an Independent in Mitchell. Though Labor’s won a landslide in this election people in South Australia have never fallen in love with Rann’s lacklustre government. He has displayed competence and has been rewarded for it, this election was more about punishing a disasterous Liberal Opposition than rewarding a mediocre government.

    Therefore if you look to where there were real alternatives to Labor you can see that these alternatives were chosen over the governing party. The Xenophon phenonomen is perhaps the greatest single example of this, while in Mitchell the vote for Hanna held up spectacularly well when Labor believed that Hanna, as a former Labor MP, would see his vote simply drift back to where it had come from. Mitchell is the neighbouring electorate to mine (Bright) and I drive through it regularly. Hanna had a significant poster presence and was known as a hard-working local MP who could get things done at the local level. He has been rewarded for this.

  8. Futher my other predictions were horribly wrong. I guessed that the seats of Hartley, Mawson and Newland would be held narrowly by the Liberals, and I predicted that Graham Gunn would lose Stuart, while Family First would be an unexpected win in Kavel. I also predicted Liberal gains in Norwood and Mount Gambier, suggesting that a big overall swing to Labor would be offset by localised factors and their majority would be stymied. I Seven wrong calls … I should hand my head in shame.

  9. A few thoughts…

    Mitchell – Hanna’s real achievement is convincing would be Liberal voters that he was the only candidate capable of defeating the Labor candidate. Accordingly there was a big drop off in the Liberal primary vote with so many voting strategically. I’m not sure if his decision to switch from Greens to Indepedenent was an act of petulance or genius, but it presumably made him a much more attractive option to those would be Liberal voters.

    Norwood – only a small swing to Labor, enough evidence of Smart pulling a personal vote. The ALP would be worried if he’s the endorsed candidate for Hindmarsh.

    Xenophon – Amazing level of support. There’s a spot in the federal Senate for him if he wants it.

  10. Just a point a little off the subject. I noticed that there were no by-elections during the last SA Parliament’s life. Isn’t this a little unusual? I’d have thought it’s quite normal for there to be at least one by-election per term … usually in the seat of the losing Leader (which was avoided as Kerin didn’t step down in 2002 and looks like he won’t be doing that this time either … most likely as his seat could be lost!), are there many other Parliaments in recent years which have been by-election free zones?

  11. Dave S, this is a subject for further research – it seems to me that by-elections have become less frequent in recent years. There have been none in the life of the current Victorian parliament, which seems pretty remarkable given that there are 88 seats and it’s been sitting for three-and-a-half years (not to mention 22 seats in the upper house which can also have by-elections, although this is about to change). On the other hand, New South Wales has been fairly fruitful recently (Pittwater, the “triple M” by-elections, Dubbo in 2004) and we’ve had three in the current parliament in Queensland (all in the past year). I would say that this is a fairly typical frequency for these states. In WA, there was the Victoria Park by-election to replace Geoff Gallop in February, but before that you have to go back to the Nedlands by-election to replace Richard Court just after he lost the 2001 election and the Merredin by-election to replace Nationals leader Hendy Cowan not long afterwards.

    But the paucity of federal by-elections lately has been really remarkable – the Werriwa by-election to replace Mark Latham early last year was the first since the Cunningham by-election of 2002, which was the only one in the previous term. A quick scan through Psephos reveals there were 140 by-elections (give or take one or two – I wasn’t counting too carefully) in the 20th century, so a year without a federal by-election is historically unusual. However, there is a particular concentration shortly after governments lose power, and perhaps also before a government suffers a predictable election defeat. So I would suggest the strong position of the Howard government explains the slim pickings in recent years at the federal level, and Labor’s big majorities in most states do the same at state level.

Comments are closed.