John Warhurst March 7, 2013 - 11:43AM
After three days in Western Sydney, the PM has yet to conduct an unscripted meet and greet in the street. Tactic, oversight, or media misunderstanding?
Kevin Rudd’s case for being recalled by the Labor Party in place of Julia Gillard is largely based on his voter appeal. He is regarded as the only leader who can win the next election for Labor or at least minimise the damage. Critics reckon Labor is mad not to change leaders because of the clear message of the public opinion polls that it will do better under Rudd.
This is the context in which the Fairfax Press reported prominently last weekend on the results of a Fairfax Media/ReachTEL automated poll taken in four safe Labor seats in western Sydney. The Sydney Morning Herald headline said it all: “Wipeout in the west: voters want Rudd”.
The poll, taken before Gillard’s western Sydney blitz, purportedly contrasted the performance of Labor under Gillard and under Rudd on a two-party preferred basis. The results were startling. The survey reported that Labor would do remarkably better under Rudd. Under Gillard, Labor would clearly lose each of the four seats, Chifley, Werriwa, Blaxland and McMahon, while under Rudd Labor would easily win Chifley and Blaxland and narrowly lose the other two.
Behind you! Prime Minister Julia Gillard and Kevin Rudd, right. Photo: Getty Images
The percentage differences were so enormous that they are hard to believe. At the very least they set new standards for leadership impact. Under Rudd, Labor purportedly did a massive 12 per cent better in both Chifley and Werriwa, 8 per cent better in Blaxland and 9 per cent better in McMahon.
There can be no doubt that Rudd is popular, but he is also an enigma. He showed his vote-winning prowess immediately he became opposition leader in December 2006. He has a phenomenal ability to attract votes. He immediately jumped ahead of John Howard as preferred prime minister and never looked back. Then in office he was personally highly popular almost up to the month, June 2010, in which he was deposed by Gillard.
He is enigmatic because his popularity in the public opinion polls is not matched among insiders, Labor MPs and public servants, who have worked with him. There is an enormous disjunction that almost defies belief. Only last week, just before these polls became public, I asked two former senior Department of Foreign Affairs and Trade officers what they thought of Rudd. They rolled their eyes. They thought Rudd was a disaster.
It is an extreme case of private disdain versus public popularity.
Rudd’s huge numbers are hard to reconcile with traditional notions of what explains voting behaviour at elections. In these traditional notions, leadership is certainly a factor but hardly as big as these polls suggest. Instead, the impact of candidates and leaders is usually outweighed by party loyalties and political issues.
I went back to the standard explanation by professors Clive Bean and Ian McAllister of Electoral Behaviour in the 2010 Australian Federal Election (in Julia 2010: the caretaker election, edited by Marian Simms and John Wanna, ANU e-Press). Their account gives due regard to leadership, as well as discussing apparently significant factors such as party allegiance and issues such as management of the economy and education.
In fact, they conclude that in such a closely fought election, ''the leadership factor was crucial''.
All other things being equal, they conclude that if the Liberals had had a more popular leader, given Tony Abbott’s greater unpopularity compared to Gillard, ''the outcome probably would have been a narrow victory for the Liberals and Nationals''. In their discussion they make it clear that leadership was a factor in Labor's favour of about 1 per cent.
Yet last weekend's poll was not suggesting an impact of this order, nor even an impact double or treble the impact at the 2010 election, but an impact of the order of 10 times Gillard’s advantage over Abbott.
There are various possible explanations. One is just that, at least in western Sydney, Gillard may be enormously unpopular and Rudd enormously popular. The gap then can be explained by the contrasting personalities alone, perhaps still fuelled by anger at the way Gillard deposed Rudd.
Another possibility is that western Sydney voters assume that a new leader will mean new government policies and that Rudd stands for something very different to Gillard. If that is so it is not clear what those differences might be. They have not been given much attention in the leadership contest between Gillard and Rudd. Nor is it clear that if change meant a higher mining tax and a more compassionate treatment of asylum seekers, that these new policies would be attractive to voters in western Sydney or anywhere else.
If a change of leader has such remarkable impact on voter choice then it also throws some doubt on a lot of other critical writing about Labor as a party in crisis.
Labor's position is generally seen as hopeless. Furthermore, there are deeper criticisms of Labor that assert not only that Labor will not win the next election, but that as a party its brand is hopelessly compromised and that it is a party without values or ethics.
This poll is not suggesting that Labor under Rudd would win the next election. These four seats are safe Labor seats after all, and even under Rudd there would still be a swing to the Coalition.
Nevertheless, there is a tendency for commentators to present Rudd’s return as a magic bullet that will make many of Labor's other problems disappear.Polls like this stretch the beliefs of political scientists. They appear to go too far. But if they are half right they still suggest the Rudd phenomenon needs further explanation. If they are wholly correct in terms of magnitude then whatever happens to Rudd, Australian politics is witnessing a quite remarkable phenomenon.
• John Warhurst is an emeritus professor of political science at the Australian National University.
