Bernard Keane was quite right to include me yesterday among pundits who got the Northern Territory embarrassingly wrong. I thought the opposition would pick up only one or two seats; instead they got four, and came close enough in a fifth to briefly raise thoughts of a hung parliament.
That might not seem a big error, but in the Northern Territory’s mini-parliament there’s a big difference between eight seats and eleven.
The two-party-preferred swing was something like 8%, which almost no serious commentator was predicting.
So what went wrong? There are basically two possible explanations:
(a) The Territory is just an exotic place, and strange things happen. Local factors, overlaid on miniature electorates, produce wild and unpredictable swings; you can’t generalise from those results to the rest of the country.
(b) This is the first move in a decisive shift of the electoral pattern Australia-wide. The fact that Labor is now in government federally has changed the dynamic and will now produce major swings to the Coalition in all the states in turn.
Evidence against (a) comes from the fact that, as I remarked on Friday, the Northern Territory has moved pretty much in line with the states over the last few elections — which isn’t all that surprising, since the exotic wide open spaces of the Territory don’t return many MPs; elections tend to be won or lost in the mortgage belt, just like everywhere else.
Evidence against (b) comes from the fact that the polls don’t seem to be picking up big swings to Labor anywhere else much — certainly not federally, where today’s Newspoll has Labor still with a 57-43 lead. (And despite occasional punditry to the contrary, we know the federal and state swings move in tandem more often than not.)
Commentators are already confidently assuring us of the truth of (a) or (b), according to their personal preferences. But the truth is we just don’t know. You don’t pick a trend from just one data point. That doesn’t mean the trend isn’t there; it might be, but it’s too early to say.
One thing we can say with confidence, however, is that voters don’t like early elections — and that could make the Western Australian election on 6 September a very interesting affair.
Poor Charles Richardson! Who is he anyway. He clearly knows nothing about the Northern Territory or its people or its political parties. His writings in recent weeks demonstrate this to the extreme.
At the time he wrote todays piece, the Country Liberals had picked up not four but seven seats – a huge difference – leaving Labor in fact with the smallest majority possible. That was a huge error Mr R. and any Psephologist worth his salt would be looking far past the “low turn-out” story (a total furphy). What this swing – and maybe the failure of many to vote, if that be true – is that LAbor had become arrogant, and had failed to deliver their well-spun promises on so many fronts. Indeed, as with Rudd today, they had become a Government by spin – not action.
What went wrong (c) 40% did not turn out to vote. Add this to the informals and there are a lot of questions to be asked by psephologists about the future of voting patterns, rather than comparisons with past voting habbits.
Thanks for the feedback.
(a) As Graeme Lewis says, the 60% turnout notion is “a total furphy”; Antony Green comprehensively debunks it here: http://blogs.abc.net.au/antonygreen/2008/08/turnout-at-the.html . Turnout will actually be close to 80%.
(b) If you go back and read my piece from last Friday, it will be clear that I’m working with the post-redistribution notional figures, under which the CLP was starting with seven seats, not four, so they have indeed gained four seats to get to eleven.