The reports of the NSW elections have been very interesting, but more for what they show about the way we tell stories about elections than for how well they explain the process and its outcome. There seems to be a struggle between two stories – a dominant one about a public drama culminating in an act of collective choice, and a subordinate one – known about, but little-used – about the structures which channel political activity. Which of these is the ‘real story’ of the elections?
The consensus of the commentariat is that the electors, by returning the Coalition to office for a third term (unusual these days), endorsed the performance of the Berejiklian government. The argument is that while there may be some unhappiness in rural areas, Sydney electors approve of the government’s record of construction, and the force of the opposition challenge was blunted in the final week by the release of the Daley Politics in the Pub speech, and by an inept interview on cable TV. Had it not been for this last week, there might have been an ALP government.
This is ‘a good story’, showing the election as the exercise of democratic choice, but it is challenged at a number of points by the ‘structuring’ story, which the commentators know about but tend to ignore. For a start, to see the election as a judgment on the government ignores the evidence that most voters vote the same way every election, and to the extent that they take notice of the government’s performance, view it through the lens of their existing party identity. This means that most electorates return the same party every time, and the only opportunity for voter choice to change the government comes in seats where party support is evenly balanced, of which there are relatively few in NSW.
Moreover, the contenders do not start from the same starting line: the government has the majority carried over from the last election, and the opposition is further handicapped by the scope for discontent with the government to be channelled to independents rather than the alternative government. This meant that the opposition would have had to get a 9% increase in its vote in order to form government, and there was no sign of this having happened: the polls showed the two sides fairly closely matched, and the betting markets (usually seen as more reliable than opinion polls) expected a Coalition victory. But the commentary presented this outcome as something of a surprise, and a crushing loss for the ALP.
Of course, in the end it comes down to a contest between a Coalition government and an ALP one, but the election is not just a two-horse race. The ‘swing’ invented by TV commentators, and the ‘two-party preferred’ figure generated by the Electoral Commission, present it in these terms, but these are artefacts of the commentator, and can be seriously misleading as an indicator of what the voters were expressing. For instance, there was said to be a 10% swing to the ALP in Camden; in fact, the Liberal vote dropped 17.5%, but the ALP vote only rose 2.5%: One Nation rose 13.1% and Others 3.8%. If there was a swing in Camden, it was from the Liberal to One Nation. In the same way, there was said to have been a 1.7% swing to the ALP in Lismore, winning them the seat, but their vote was exactly the same as 2015; the Nationals vote shrank 2.5%, and the Greens 2.1%, and an independent candidate won 5.1%. Is this a swing to the ALP, an endorsement of the government, a critique of it, or a reflection of the fact that the Nationals’ MP was retiring after 20 years in the seat, and there was an attractive alternative?
The commentators, like the government, want to see the election as a single two-horse race, but it was made up of 93 separate races. This is the problem with the widely-believed story that Daley’s pub statement cost the ALP support among voters of Asian background. The only evidence offered is that ALP vote in Kogarah fell 2.9% and the vote for the Liberal candidate, who was of Chinese origin, rose 8.3%. But in neighbouring Heffron, the ALP vote fell 2.4%. the Liberal cote fell 2.1% and the Greens vote fell 2.3%, Keep Sydney Open got 8.7% and Animal Justice rose 3.1%. Next door in Coogee, the ALP vote rose 2.6%, the Liberal vote fell 5.2% and the Greens vote 4.5%, and Keep Sydney Open gained 4.5%. And in Blacktown, where 25% of the electorate is of Asian origin, the ALP vote rose 1.6%, the Liberal vote fell 6.1% and the One Nation vote rose 6.9%. These figures prove nothing, but they are not evidence of an Asian backlash. What was distinctive about Kogarah is not that the Liberal candidate was Chinese, but that he had just been featured in Chinese state television in a program on Chinese who had ‘done well’ overseas; why was this less significant than what Daley had said in a pub in the Blue Mountains a few years ago?
Both the ‘voter choice’ and the ‘structuring political activity’ stories have their uses. The ‘voter choice’ story is used to validate the outcome – who forms government, and what a re-elected government has done, or does subsequently. The claim that judging the government is what the voters did is a myth, but a useful myth: it justifies the position of the political leaders, and it facilitate change in the way that we are governed. It is a ‘front-stage’ story. The ‘structuring’ story is a ‘back-stage’ story, shared by the operatives, but less talked about. It is less worried about justification than the dynamics of governing: what people do (and don’t do) and what difference it makes. The commentators are usually operating front-stage and their accounts are framed in ‘voter choice’ terms, but they know the ‘structuring’ account, though they’re never quite sure how much of it to acknowledge in their own stories. If you want to talk about the game, you have to accept its own story.
H.K. Colebatch is a political scientist who established the graduate program in social science and policy at UNSW. But he has not published any poetry, nor has be been awarded the Prime Minister’s prize for Political History.