Sean Kelly
The two battles of the 2016 election

Perhaps the greatest blow in modern times to the reputation of human prognostication occurred about a month ago, in the United States, as the realisation sunk in that, contrary to what pretty much every expert of any standing in the country had predicted, Donald Trump had won the Republican presidential nomination.

What followed was a desperate and unedifying race between pundits to concede their errors – best to get in front of these things – and explain how they had made their mistakes. And oh, what explanations. They were complex, they were arcane, and most of all they were abundant. But amid these many and varied reasons was a simple and fatal error, and it’s interesting in the context of Australia’s own election.

We complain a lot about polls in this country, and fair enough. We are snowed with them, and without wanting to rehearse tired arguments, accurate as they are, it’s fair to say they’ve had an invidious effect on politics during the past decade. But the interesting thing about the Trump case is not that polls received too much attention. In fact they received too little. Political analysts by and large ignored consistent evidence showing Trump was caning his competitors. As Nate Silver, the stats whiz who became famous for predicting Obama’s 2008 victory with precision, but who consistently called Trump wrong, ruefully concluded on Twitter: “Roughly speaking (I’m generalising a LOT) polls got Trump *right* while other types of empirical evidence (e.g. endorsements) got him wrong.”

Australian polls, including both Newspoll and Essential this week, show Labor and the Coalition tied for the national vote. Dead heat, 50-50. And yet voters continue to expect a Liberal victory. The bookies even have Labor drifting further out of contention. The reasons for that disconnect tell us a lot about this election.

The first, which everybody’s heard a thousand times by now, is that elections are won in marginal seats. Political strategists on both sides are telling journalists that Labor may have an even share of votes nationwide, but is behind in the seats that matter. This is probably true. Incumbents have an advantage – they’re better known – and the government, by definition, has more incumbents. Still, it’s hard to be certain. Even relatively small-scale polling is expensive and campaign funds are scarce. To be accurate, marginal seat polling has to canvass thousands of voters, and is not done as often as you might think.

The second reason is that an idiosyncrasy of our electoral system generally means Labor has to win a little more than half the vote to win a majority of seats, as Kim Beazley and Julia Gillard both learnt in different hard ways.

But the final reason reveals the most. Late last week the Liberal Party’s pollster, Mark Textor, in a podcast with Guardian Australia, explained that voters today are different from voters 20 years ago. In the Media Saturation Age, people understand, when they are polled, that a poll will be published, and that it will have an impact on the media cycle. They are therefore capable of telling the pollster one thing while intending to vote a different way. “People game the polls…” he said. “That is: ‘I’m not really going to vote against Labor or Liberal when the election comes around but I want to give them a tickle-up based on a particular issue I have today.’ What the poll’s also failing to do is pick up the rise and rise of what we call tactical voting, that is having a feeling about a political party at one time but having a more mercenary approach to your vote at the end of the day”.

Asked to explain precisely what might make people say one thing and do another, Textor answered: “Well, there are many possibilities and no one wants to telegraph strategy and it’s not a strategy but one of the determining factors of this election is the extent to which they want stability in government and stability in the economy. Now, what we’re saying quite rightly and what is a very strong argument for the Liberal Party is that when you have a change in government you have enormous disruption … Voters can have the choice these days in the knowledge that they have the power to overcome their feelings about a particular party or their approach on any given day and in a more mercenary sense or a more realistic sense decide to vote on some meta-basis: for stability, or for fairness, or for something else.”

And that, ladies and gentlemen, is all you need to know about election 2016. It is a battle of two battles. The Coalition needs voters to see the election as a choice between stability and instability. The ALP needs the same voters to decide their vote based on fairness versus unfairness.

Don’t take my word for it. Lazarus himself, John Howard, who can always be counted on not only to return from political slumber but to deliver the political goods, was this week campaigning in the usually safe Liberal seat of Mayo, and told Adelaide radio: “I think what the public wants is a bit of stability, and if you want that there’s only one bloke to vote for and that’s Malcolm Turnbull.”

Howard was helping defend Mayo from Nick Xenophon’s candidate, which was interesting given that the headline result from this week’s Newspoll was the rise of “independents and others” to their highest vote during a formal campaign in the poll’s 31-year history.

Asked about that result, the current PM responded that “a vote for Labor, a vote for the Greens, a vote for independents risks the chaos and the instability that we saw in the years of Julia Gillard’s government. A stable Coalition government, which I lead, will deliver on our national economic plan and will deliver strong growth and more jobs.”

This is what a smart campaign does: every negative mentioned should reinforce your own strengths. The Coalition’s repeated warnings of the dangers of a hung parliament are only partly about a hung parliament. Mostly, they should be read as another way to remind voters that Turnbull is Father Stability, while everyone else is Captain Chaos.

This came through in Turnbull’s Wednesday night interview with Leigh Sales, too, in which he made the case – slightly mournfully, it seemed to me – that “my government’s economic plan is working, but I need three more years to complete it and that’s what I’m seeking from the Australian people.” This is the same strategy NSW Labor employed to re-elect premier Morris Iemma after he had replaced Bob Carr, running on the slogan “more to do but … [we’re] heading in the right direction”.

To win, Turnbull must not only get voters to focus on stability. He must neutralise concerns about fairness. This was the thinking behind a new ad presenting Turnbull’s past, growing up without a mother and without the millions he would make later in life. Labor’s brutality towards top hats has clearly had an impact, which you can see in Newspoll, which now has Turnbull at his lowest net satisfaction since getting the job, and in Essential, which found that 33 per cent of voters say they have developed a less favourable view of the prime minister during the campaign. Only 7 per cent have improved their opinion.

Labor must do the opposite, getting voters to focus on fairness while neutralising concerns about its own ability to deliver stable, unworrisome government. And so Bill Shorten and his treasury spokesman Chris Bowen released Labor’s 10-year economic plan on Wednesday. Economics, along with boats, is the strongest political proxy for reliability.

Both parties now have 10-year plans for growth. Both plans have six parts to them – a number strategists must believe seems at once specific and sufficiently large. Both plans aim to shrink the deficit year by year, and to deliver a surplus in 2021.

The difference between the two is that Labor’s plan shrinks the deficit more slowly in the first four years, then speeds up after that. That may sound like a wonkish detail, but the party is taking a huge political risk. Four years is the standard by which budgets are generally judged. Turnbull and Scott Morrison will hammer this until election day. They are not entirely without sin, given much of their own plan for growth is dependent on the company tax cut, which does not have much impact for a decade. But that won’t stop them. Labor will have to decide whether to ignore the attacks, hoping the caravan moves on, or counterattack with ruthlessness, loudly demanding the government provide 10-year numbers, too.

Labor’s hope is that it has done enough on the stability front and more than enough on fairness. At the weekend it released its childcare policy, squarely in the fairness space. Having dragged the campaign onto economic grounds it will want to return to fairness, or to policies that do a little of each, such as the national broadband network. Labor’s attacks on the company tax cut – their great symbol of unfairness – suffered a little when video emerged of Shorten backing a similar cut in 2011. But much like the Coalition’s vulnerability on 10-year time frames, I don’t think this will hold Labor back from here on in.

The candidates took some time out of campaigning this week to visit flood-affected regions. That calm will vanish next week as the candidates meet for probably the final debate of the campaign, which will be broadcast on Facebook, on, and perhaps on television. It might be the best chance the leaders get to speak directly to voters. The words they use might vary, but the language they’ll be speaking will be clear: stability and fairness, all the way.


Paul Bongiorno is on leave.

This article was first published in the print edition of The Saturday Paper on June 11, 2016 as "Saying one thing, doing another".

A free press is one you pay for. Now is the time to subscribe.

Sean Kelly is a columnist for The Age and The Sydney Morning Herald, and a former adviser to Labor prime ministers.

Sharing credit ×

Share this article, without restrictions.

You’ve shared all of your credits for this month. They will refresh on July 1. If you would like to share more, you can buy a gift subscription for a friend.