Queensland Labor experienced two major electoral events in 2004: a landslide win in the February state election, which was surpassed in modern history only by the triumph of 2001; and a failure at the October federal election so immense that it handed the Coalition control of the Senate. Even after the monster swing at the 2007 federal election, which netted Labor nine seats, the federal Labor vote of 42.9 per cent was still well short of the 46.9 per cent polled by Beattie in September 2006. While other states duplicated this pattern to some extent in the era of Howard government and state Labor hegemony, nowhere else did state and federal voting become so spectacularly uncoupled.
Why this might be so is a discussion for another time: the purpose of today’s lesson is to examine regional variations in state-federal vote splitting. This can hopefully give us a sense of which areas have been over-performing for state Labor, and are thus ripe for correction – remembering that we might equally be measuring where Labor was over-performing federally. For the purposes of this exercise, I have divided the state into seven regions of variable size and ascertained the parties’ primary vote in each at two “pairs” of elections: the state election of September 2006 together with the federal election of November 2007, and the state election of February 2004 with the federal election of October 2004. The number of state districts in each region is indicated by the numbers in brackets. The top end electorate of Cook has been left out of the analysis because it wasn’t a good fit with either of its neighbours.
The results in turn:
Brisbane. I arguably should have divided this into smaller regions, but I suspect each would have told a similar story. The figures confirm what we already knew: that any conceivable change of government will involve a huge swing and a sackload of new seats for the LNP in Brisbane.
Gold Coast. This area is home to five Labor MPs out of eight at state level, whereas the Liberals hold the three corresponding federal seats by between 13.9 per cent and 19.5 per cent. The Labor gap narrowed only slightly between the two pairs of elections: Labor actually picked up a slight swing at the last state election, whereas their swing at the federal election was 2 per cent lower than the rest of the state.
Sunshine Coast. Defined as the pre-distribution seats of Caloundra, Kawana, Maroochydore, Nicklin and Noosa. The one area in the state where the conservatives made satisfactory inroads in 2006 (aided by water issues which haven’t gone away), the Sunshine Coast also swung by about 8 per cent to Labor at the federal election, resulting in by far the biggest narrowing of any region. All five seats are now safely in Liberal or, in the case of Nicklin, independent hands.
Hinterland. Covering Beaudesert, Cunningham, Glass House (possibly should have put that one in Sunshine Coast), Lockyer, Nanango and Toowoombas North and South. This area has behaved fairly typically of late in swing terms. The only Labor seat is Toowoomba North, with a margin of 7.6 per cent. Glass House is a Labor-held seat in a sense, but the margin is literally non-existent and it has been deserted by its sitting member, making it an almost certain win for the LNP.
Central Coast. From Gympie north to Whitsunday. This region swung modestly to the Nationals in 2006 and heavily to Labor in 2007 (it includes the southern two-thirds of Dawson, where the overall swing was 13.2 per cent), which narrowed the gap substantially. The low-hanging fruit here is Whitsunday, Hervey Bay and Mirani (the latter of which is only notionally Labor); further up the tree is Keppel with a margin of 8.1 per cent.
Northern Coast. Burdekin north to Barron River, including Townsville and Cairns. This is the only region where the gap widened between the two elections (it might have been otherwise if I had included Cook, due to Labor’s exceptionally strong swing federally in Leichhardt), partly because Labor’s vote doubled in Hinchinbrook in 2006 after an independent performed strongly in 2004. Even so, Labor performed exceptionally well across this area in 2006, and margins in seats such as Barron River (4.8 per cent), Cairns (8.0 per cent), Townsville (9.4 per cent) and Mulgrave (9.8 per cent) would be as inflated as those in Brisbane, if not the Gold Coast.



6 Comments
William, it is unlikely that this election will be the one where any major correction occurs
towards the Nationals. More likely at this stage is that multiple seats held by the Nationals are quickly moving towards independents. The baseball bats are out and the independents are the ones doing the swinging.
Let me just say Billbowe, they are very pretty tables!
There was a lot of regional variation last time, and it’ll be interesting to see if that occurs again.
FWIW, a lot of the swing to Labor in Central and North Queensland (and in seats between Ipswich and Toowoomba) in both 2006 state and 2007 federal is attributable to the same factor – WorkChoices. Labor ran very heavily on it in 2006 in electorates with high numbers of casual employees and low median incomes.
OK William,
I’ve been staring at these numbers all night, trying to work out what you are trying to tell me. I don’t get it.. can please I ask some questions?
What caused such a tight clump of figures (- 15.2 >>> -19.6) in 2004 LNP Difference?
Given this range covers seats totalling 82 out of 89 seats, how does the average come out at -11.2%. Are all 82 seats outweighed by the 7 interior seats?
Ooooo – I appear to have acquired an avatar without trying…thankyou ..but….it looks like I sat on a pencil
Mr Squiggle, the fact that I stuffed up my Brisbane federal calculation for 2004 probably has something to do with it. Good to see somebody’s paying attention. Now corrected, although the regional difference figures from 2004 still look a little high compared with the state result. There might be a logical explanation for that – I’ll have a closer look when I get time.