Time to learn the many lessons from a long campaign

It may have taken just over a week but the result of the Australian election is now decided, with Malcolm Turnbull’s Coalition to form government. But the lessons and insights from the extraordinary 2016 national poll will be debated for years among political scientists and campaign experts.

Like a grand social experiment, the 2016 federal election provided the chance to verify, refute or validate some of the key hypotheses of Australian politics and the strategies of the campaign playbook.

With 95% of Australia’s 15.7 million enrolled voters having chosen between 57 parties and 1625 candidates to elect 226 members to the federal parliament, here are some of the key things we have learned from the 2016 election.

The Australian vote is fragmenting

Arguably the most far-reaching dimension to the 2016 election result is the decline in the vote for Labor and the Coalition parties (the “major parties”).

With some notable exceptions, the vote for minor parties and independents (“other parties”) has been trending upwards since the 1950s, accelerating from the 1990s, and reached a high of 23% in this election. In South Australia, the popularity of Nick Xenophon lifted the primary vote for the non-major parties above Labor’s for the first time.

The rise of minor parties is driven by two factors that are unlikely to reverse.

The first is declining party loyalties – the number of voters who identify as a supporter of a political party is falling and voters are more likely to change their vote from one election to another creating increased volatility in election results.

Second, political divisions have become more complex, straddling economic, social, cultural, geographic and other issues in ways that do not line up neatly with the major parties. In the same way consumer product markets have fragmented so to have voting markets meaning people are more likely to vote for a niche party or candidate rather than one of the broad mainstream parties.

In the 1950s you were either a Ford family or a Holden family. Now there are more than 40 car makers in the market and literally hundreds of models. It is a similar story in the beer market and many others.

In this election, voters could choose from a record number of 57 parties. On one level, it is remarkable that the vote for the major parties in Australia has remained as high as it has. But the experience of other advanced democracies internationally would suggest the trend is set to continue.

This will have a profound impact on Australian politics. To begin with, it means the 45th parliament will have the biggest and most diverse Senate crossbench since Federation. Recent experience suggests this will make the passage of legislation difficult.

In the House of Representatives, minor parties and independents have won five out of 150 seats. The domination of the major parties is preserved by the preferential single-member voting system which translates a 77% share of the vote into 97% of seats in the chamber. But even with the current voting system we can expect minority governments to become more prevalent, possibly even the new normal.

It also means that the preference flows of minor parties and independents are now a critical factor in determining who forms government. It begs the policy question of whether the preference deals that are negotiated between the parties should be the subject of increased public scrutiny, beyond the tabloid treatment they receive at the moment.

The media is fragmenting

The decline in the influence of traditional media in modern elections is well documented, and was more pronounced than ever in the 2016 election.

A fragmenting media market means no single media source or outlet has a dominant influence over the electorate. It also means that the political parties are bypassing traditional media and using social media and direct channels to communicate with voters.

An example of the shift in media influence was the 2016 campaign run by the News Corp press that strongly backed the Coalition. The Daily Telegraph published a string of highly partisan front pages attacking Bill Shorten with satirical cartoons.

It is hard to discern the influence of this bias given we do not know what the election result would have been without these interventions. But it clearly failed to stop a very large swing to Labor in western Sydney.

The political parties know the role of traditional media is diminishing and increasingly prioritise direct one-to-one contact with voters through social media, emails, phone calls and door-knocking. The Liberal Party harnessed the power of popular Chinese-language social media platform WeChat to reach voters in the ethnically diverse seat of Chisholm in the eastern suburbs of Melbourne.

The Greens used Grindr, the world’s largest gay social network app, to reach voters in the Victorian seats of Batman and Higgins.

Facebook remains the king of social media in Australia, and this is where the major parties put most of their efforts. Data collated by social media tracking tool CrowdTangle shows that over the eight-week campaign Labor’s Facebook page had 708,152 interactions on its page, while the Liberal Party’s page had 341,520.

In the 2016 campaign, Labor identified 1.6 million voters who were persuadable in marginal seats. These voters were targeted through social media and phone calls with pro-Labor campaign information and structured conversations on issues of interest to the individual voter.

With the mainstream media audience seemingly more interested in My Kitchen Rules than electoral politics, this peer-to-peer communication has rapidly grown in importance to campaign strategists.

This development is further accelerated by a Canberra press gallery that got the forecast of the poll result and the key issues of the election spectacularly wrong. The echo chamber of the gallery concluded that Turnbull was going to win with a comfortable but reduced majority, and this influenced the way the election was covered.

The endless commentary about Turnbull’s personal popularity being decisive, the costings of the party policy commitments and the impact of the UK’s departure from the European Union all missed the mark. Opinion polls and social research show that the campaign staples of “health, education and jobs” were once again at the fore of voters’ minds in 2016.

With the press gallery not talking about these issues the parties now bypass the media to have conversations directly with voters about them. This should be cause for some serious reflection by those reporting on the election.

Long election campaigns work against an incumbent government

The Australian political playbook holds that long election campaigns hurt incumbents. Turnbull set out to test this theory by effectively running a record 12-week election campaign.

Turnbull officially announced the election on May 8, but in reality he announced it at a press conference on April 19 when he outlined his plan to reintroduce the industrial relations bills that would be a double-dissolution trigger.

The conventional wisdom is that a long campaign disadvantages the incumbent by giving up many of the advantages of governing. In an election the government and opposition are on a level playing field, with the opposition given close to equal media coverage. The elevated status of the prime minister is diminished as the government moves into caretaker mode and the prime minister becomes just another politician trawling the shopping centres of Australia for votes.

Bob Hawke was the last prime minister to run a long election campaign in 1984. The then-opposition leader, Andrew Peacock, is considered to have run a good campaign with a surprising 2% swing against the government, costing it nine seats.

Nearly 30 years later, the Liberal Party thought the long campaign would expose the shortcomings of Bill Shorten and the Labor opposition. Instead the Coalition suffered a 3.5% swing and lost between 13 and 15 seats, subject to final counting.

On one level, the published polls did not move significantly over the period of the long campaign. But Turnbull emerges from the process greatly diminished, and an abiding sense that the government squandered the benefits of incumbency with the long campaign. So the rules of the campaign playbook still stand, and we can expect it to be another 30 years before we see another long election campaign.

Next: Great scare campaigns in electoral history