That we’re going to an election in September was no surprise. But the date doesn’t make Julia Gillard’s road to retaining government any easier. Crikey’s poll cruncher looks at what she has to do to hold on.
Diverting though yesterday’s election date announcement may have been, its contribution to the sum of human knowledge should not be overestimated. There was already no remaining doubt the Prime Minister was budgeting for a normal full-term election for the House and half the Senate, which left her with a fairly narrow window of opportunity for its timing.
The unorthodox alternative of an earlier poll exclusively for the House of Representatives was, and remains, foreseeable only in the event of parliamentary rebellion or leadership change. These are distant prospects, but they haven’t been made any more so by Gillard announcing when she will, God willing, hold the election.
Nonetheless, the occasion provides a useful focusing moment to survey the electoral terrain ahead of what some would characterise as an eight-month election campaign.
Polling aggregators having proved their worth at the US presidential election last year; a number have sprung up in Australia with a view to ironing out the biases and volatility that apply when following an individual poll series. Whatever their methodology, each says much the same thing: the Coalition has a polling lead of around 52.5-47.5, and has done so since a trend to Labor levelled off about three months ago.
But published polling offers at best a limited guide as to how the swing might be distributed geographically. The best resources available are the mainland state breakdowns featured in Newspoll’s quarterly aggregates and the regular monthly results from Nielsen.
The former appear infrequently, are partly out of date when they arrive, and are subject to the fairly luxurious margins of error which apply to individual poll results even when they come from solid samples. Nielsen doesn’t wait to accumulate three months of data before publishing state-level results, which means its samples for individual states are in some cases vanishingly small.
And aggregated measures of the two pollsters do not sit comfortably with the consensus inside the beltway, which is presumably informed by better targeted internal polling. Here the view is that Labor faces potentially insurmountable difficulties in Sydney, with none of the 11 seats held on margins of less than 10% to be considered bolted down. However, Newspoll and Nielsen furnish no evidence of an unusually severe swing in New South Wales, if anything showing slightly bigger swings (off a much higher base) in Victoria.
A poll did finally emerge earlier this week which went according to script. Conducted by automated phone pollsters JWS Research for a Liberal-linked client, it pointed to a collective double-digit swing across the state’s marginals.
Labor’s other mooted disaster zone is Tasmania, which appears headed for one of its occasional realignments (the Liberals having held all five seats from 1975 to 1987, and Labor doing so more often than not since 1998). Here evidence is even scarcer, as neither Newspoll nor Nielsen publishes results for the state owing to prohibitively small samples. But what little polling has emerged — like last week’s ReachTEL result pointing to a crushing Liberal victory in Bass — has supported the view that all four of Labor’s seats are at risk.
“The fundamental task for Labor over the next six months is to claim ownership of the strong economy …”
Even leaving polling aside, the consensus view has much to offer it intuitively, particularly if it is accepted state factors are likely to play their part.
In NSW, the voluminous dirty laundry of the previous government continues to receive a daily airing via the Independent Commission Against Corruption, while disappointment with Barry O’Farrell remains within acceptable limits. In Tasmania, a Labor government that had reached the end of its natural life by the March 2010 election has governed since in coalition with the Greens, activating hostility among blue-collar workers outside Hobart and providing unhelpful parallels with the Gillard government’s minority and Greens-supported position in Canberra.
The other side of the challenge for Labor is winning new seats in the more welcoming environments of the other states, which they will need to do even on the happiest of scenarios in NSW and Tasmania.
The technicality of Craig Thomson’s independent status aside, Labor holds 72 seats in a chamber of 150, and has next to no chance of repeating the coup of having the votes of New England (Tony Windsor) and Lyne (Rob Oakeshott) in its column. The most promising prospect for covering the gap is Queensland, where the travails of the Newman government have helped Labor claw back to something like its 2010 election position less than a year after appearing headed for total annihilation.
Elsewhere around the country, battleground seats are relatively thin on the ground.
Labor will have its work cut out defending a number of marginals in and around Melbourne, but the election is nonetheless likely to continue a trend going back to 2007 of Victoria assuming lesser tactical importance than its northern neighbours. Sydney, Brisbane, Tasmania and the central coast of NSW may be the crucibles, but it doesn’t follow that Labor can build a strategy out of sandbagging marginals and surgically targeting Coalition weak spots.
Raising the boats will require a higher tide, which means an improvement in Labor’s position nationally.
The fundamental task for Labor over the next six months is to claim ownership of the strong economy, promote a cohesive positive message from government policy initiatives and make hard political capital out of the electorate’s palpable disquiet about Tony Abbott.
Unless these prerequisites are met, a well-targeted ground campaign can’t hope to achieve any more than limiting the damage.