After Losing 30 Newspolls, can the Coalition be Written off?

Monday’s Newspoll is the latest in a long string of polls putting the government in an election-losing position, a trend that has continued since shortly after the last election. But is it game over for the Coalition? What do the polls really tell us? And can we rely on them? Here, we we take a look back at polling history to see what we can judge …

Does the 30th Newspoll result mean the Coalition will lose the next election?

While the Coalition has been consistently behind in the polls, recent history suggests that Labor’s lead is not big enough to rule out the possibility of a Coalition recovery, with the lead not as large as in past elections.

Image result for After losing 30 Newspolls, can the Coalition be written off?

Labor has consistently led in the polls since the last election, mostly with a two-party-preferred figure between 52% and 55%.

Labor’s lead has been relatively consistent, which could mean that many swing voters have locked in their vote. This is a strong position but it’s not strong enough to write off the government.

Would a change of leader make any difference in the polls?

Changing leaders has become a standard reaction of a party facing the prospect of an election defeat but these shifts can be illusory. In every case, support for the new leader gradually declined, sometimes back to an election-losing position.

If there is a change of Liberal leader, we could expect a shift in the polling but this may be only temporary. It would take time before the impact of the change on the polls became clear.

The Abbott-led Coalition government trailed Labor for most of 2014 and 2015 but the Coalition jumped back into the lead when Malcolm Turnbull succeeded Abbott in late 2015. Yet this lead did not prove stable. Labor gradually regained ground as the 2016 federal election inched closer, and the two parties were close to a tie by election day.

Big shifts also occurred in the years before the 2004, 2007, 2010 and 2016 elections, generally after a change of leader by one ot the major parties. Opposition parties gained ground after installing Mark Latham (2003), Kevin Rudd (2006) and Abbott (2009), while governments have recovered from poor polling under new leaders Julia Gillard (2010), Rudd (2013) and Turnbull (2015).

Image result for After losing 30 Newspolls, can the Coalition be written off?

Has any other government suffered such a sustained slump in the polls, but gone on to win?

There’s no perfect comparison, but we have definitely seen governments come back and win after falling behind. The Howard government trailed the opposition in most of the year leading up to the 1998 and 2004 elections, but recovered and won in both years. Neither of these poll slumps lasted as long as the current government’s slump.

Of course, the current government did recover following the change to Turnbull’s leadership, and we don’t know if the same trick could work a second time.

How reliable are the polls?

Australian polls have done a reasonably good job of predicting national results (although individual seat polling is much less reliable).

Polling averages before the 2016 election accurately predicted a narrow win for the Coalition, and polls were similarly close in predicting the national result at every election since 2007.

The last election where polling failed to predict the result was in 2004, when last-minute polling suggested a tight race, but the actual result was a comfortable win for the Coalition.

Of course, current polling is not a prediction of the election outcome, and there is always room for error, but current national polls do a pretty good job of predicting the final national result as election day nears.

What about ‘preferred prime minister’ – does that matter?

Despite leading on the other standard question asked in national polling – “preferred prime minister” – the government can’t take much comfort from it.

Turnbull continues to outperform Bill Shorten when voters are asked to name their preferred prime minister but this question does not do a good job of predicting election results. There is a substantial bias towards the incumbent, and opposition leaders in an election-winning position often remain behind on this question until they win power.

Satisfaction questions also do not do a particularly good job of predicting an election, though they can be helpful in explaining a party’s polling position. Turnbull had a strong positive satisfaction rating when he became prime minister yet this crashed into negative territory in early 2016, and he has remained neck-and-neck with Shorten pretty much since the last election. Party leaders are often replaced at a point where they have poor satisfaction ratings.

The current polling paints a clear picture: Labor has a longstanding solid lead, but it is not large enough for the government to give up hope. Personal satisfaction ratings for Turnbull are poor, which could encourage some within his party to consider an alternative leader.

It’s worth comparing the latest polling with support for Tony Abbott and the Coalition in opposition before the 2013 election. Labor scraped back in to government as a minority at the 2010 election and, within months, the Coalition had taken the lead.

The Coalition’s lead went up and down throughout 2011 and 2012 but the Abbott-led opposition tended to hold a bigger lead, with some polls putting the Coalition on more than 60% of the two-party-preferred vote. Its support dropped as the election neared, and the result was closer than the polls had predicted for most of the previous term.

What about ‘preferred prime minister’ – does that matter?

Despite leading on the other standard question asked in national polling – “preferred prime minister” – the government can’t take much comfort from it.

Turnbull continues to outperform Shorten when voters are asked to name their preferred prime minister but this question does not do a good job of predicting election results. There is a substantial bias towards the incumbent, and opposition leaders in an election-winning position often remain behind on this question until they win power.

Satisfaction questions also do not do a particularly good job of predicting an election, though they can be helpful in explaining a party’s polling position. Turnbull had a strong positive satisfaction rating when he became prime minister yet this crashed into negative territory in early 2016, and he has remained neck-and-neck with Shorten pretty much since the last election. Party leaders are often replaced at a point where they have poor satisfaction ratings.

The current polling paints a clear picture: Labor has a longstanding solid lead, but it is not large enough for the government to give up hope. Personal satisfaction ratings for Turnbull are poor, which could encourage some within his party to consider an alternative leader.

If there is a change of Liberal leader, we could expect a shift in the polling but this may be only temporary. It would take time before the impact of the change on the polls became clear.

Source