Opinion

Richard Denniss
Money, votes and the ‘pendulum’

What if money didn’t matter much in Australian politics? Clive Palmer just spent $53 million on ads for his United Australia Party and had zero candidates elected. The Greens spent about $320,000 and elected six senators and Adam Bandt in the lower house. Voices for Indi, after electing Cathy McGowan in 2013 and 2016, just made history when they elected Helen Haines, making her the first independent to succeed another independent in the house of representatives. Their budget was minuscule compared with the major parties. And Jacqui Lambie won a senate seat having gone through a lot of shoe leather but spent only $50,000.

So Palmer proved you can’t “buy an election” – even if you spend more than both major parties combined. And poorly funded, well-respected local candidates are winning more lower house seats than at any other time in modern Australian politics. Money matters in Australian politics, but by no means do those with the biggest chequebooks always win. Ironically, exaggerating the importance of money in Australian politics helps empower those with the most money to spend.

Imagine if the major parties were neck and neck in the national polls and their local candidates had similar appeal and similar voter support. In such a situation, there’s no doubt an extra hundred thousand dollars in campaign contributions could be decisive. But would $100,000 close the gap between a poor candidate and a great one? Would Tony Abbott have avoided the 12.6 per cent swing against him if he spent an extra million dollars? Or $2 million? I doubt it would have helped at all.

So if $53 million can’t buy a single seat, why do ministers debase themselves and sup with those willing to pay a few grand for the privilege of watching them eat? And why do political parties find it so hard to refuse donations from the tobacco, coal and gambling industries, even though in doing so they risk the wrath of voters by taking such tainted funds?

There is a symbiotic relationship between corporate donors and those responsible for fundraising in political parties. The corporates who donate a few hundred thousand dollars, often to both political parties, are pretty much assured good access to frontbenchers in the years to come. For a company such as ANZ, which gave $250,000 to both major parties in 2018, these donations are a rounding error compared with an annual advertising budget of $200 million. Big donors don’t always get the legislation they want, but they always get the meetings they want.

Australia is one of the richest countries in the world. According to the World Bank, our GDP is the 13th largest in the world, just behind Russia and South Korea. We are in the G20, the OECD and the Five Eyes intelligence alliance with the United States, Britain, Canada and New Zealand. We are a big and important country, but apparently a $5000 dinner or a $250,000 donation can buy you the ear of our leaders. How embarrassingly cheap our politicians must seem to US companies accustomed to handing over tens of millions of dollars. It happens because of an extraordinary desire to fundraise and a misplaced belief that money, rather than good candidates, can decide outcomes.

Everyone knows that Australia’s political culture is broken. Property developers are major donors to the politicians who rezone land. Our former defence minister, Christopher Pyne, has flouted the ministerial code of conduct and moved straight from parliament to advising defence companies on how to win government business. But while donation reforms and other accountability measures might help, the real problem is the willingness of MPs to act in such ways and a political culture that accommodates rather than excoriates such behaviour.

The “electoral pendulum” is broken, too. The metaphorical arm is supposed to swing according to the national mood and the further it swings the more seats it knocks over. But there is no “national mood” anymore because in modern Australia not only do voters in different states swing in different directions, so, too, do voters in neighbouring electorates. When local factors and local candidates matter, national pendulums and predictions don’t.

Most national political commentary is based on national averages: average unemployment rates across the country, average cost-of-living increases and average swings to political parties. But no one lives “across the country”; they live in a city or town with local economic and political conditions.

While talking about national averages makes national commentary easier and cheaper for media outlets, averages often conceal more than they reveal. The average Australian, for example, has slightly fewer than two legs as the number of one-legged people is smaller than the number of three-legged people and the number of people with no legs is larger than the number of people with four legs. Just as it is meaningless to describe Australians as having an average of 1.99 legs, or the average woman as having 1.7 children, it is now virtually meaningless to talk about the “average swing” across Australia.

In the 2019 federal election, Labor had swings to its primary vote in Victoria (1.3 per cent), South Australia (3.8 per cent) and the Northern Territory (1.9 per cent), but it won no extra seats in those jurisdictions. At the same time, it had swings away from it in Queensland (4.2 per cent), Tasmania (4.3 per cent) and New South Wales (2.4 per cent), where it lost a combined five seats and gained one. Labor also had swings away from it in Western Australia (2.7 per cent) and the ACT (3.2 per cent) but lost no seats there. All of this context is lost in the factually true but electorally irrelevant statement that Labor suffered a 1.4 per cent swing against it nationally.

The metaphorical idea of the “electoral pendulum” is what has trained us to focus on “the national swing” and its impact on “marginal seats”. Yet when the vote for parties is rising in some states and falling in others, talk of national swings becomes more of a distraction than a prediction. When good local candidates can draw thousands more votes than the swing to their party would predict, the whole idea of marginal seats loses meaning. Tony Abbott’s seat of Warringah was “safe” until a great candidate came along.

In the 2019 election the Liberal–National Coalition won only three seats out of the six predicted by the pendulum to go their way. The Coalition also won two seats the pendulum said they would not win. The ALP picked up one.

Put another way, the pendulum was wrong about twice as many seats as it was right about. On top of that, the pendulum had nothing to say about the Coalition losing a seat to an independent in Warringah, and winning one back off an independent in Wentworth.

So what’s going on? Why don’t electorates fall in the neat lines predicted by the pendulum anymore? The first part of the answer is that the pendulum was never perfectly predictive. Back in 2001 the pendulum predicted half of the seats that changed hands. It was never terribly accurate, but its accuracy is in steep decline.

Second, the decline in the number of people watching the TV news and reading broadsheet newspapers means that fewer people now consume “the national media”. In an environment in which everyone is consuming different media, the idea of “the national mood” loses meaning, and, as we saw in the 2019 election, so, too, does the idea of “the national swing”.

Third, the rise of social media means that backbench MPs, minor parties and new candidates can engage directly with their potential voters, in real time, at zero cost. While low-cost campaigning techniques such as standing on street corners and doorknocking were always available to candidates, engaging through social media is a lot quicker, more targeted and in turn more productive for a hardworking candidate. Social media also means candidates can buy much cheaper and better-targeted advertisements than they could in broadsheet newspapers or on television.

Fourth, as trust in political parties declines, the integrity of candidates matters more and more. In the 2019 election, a number of high-profile Liberal candidates including Tim Wilson and Dave Sharma ditched the Liberal Party’s official branding and went with their own. And when Ged Kearney replaced David Feeney as the candidate for Batman (now Cooper), what was a marginal seat Labor feared the Greens might take became one of the safest ALP seats in the country. Imagine how much money the ALP has saved by Kearney’s profile and stature in the community.

The popularity of candidates seen by their electorates as authentic is not confined to Australia. In the US Alexandria Ocasio-Cortez was working as a waitress when she challenged the chair of the Democratic caucus, 10-term congressman Joe Crowley, and defeated him comfortably. Since her unlikely victory she has shot to international prominence and resuscitated interest in the Democrats among a large number of progressive voters. AOC, as she is often called, once explained her success against the Democratic establishment by saying: “You can’t really beat big money with more money. You have to beat them with a totally different game.”

This all adds up to the fact that the quality of candidates running for parliament matters more than it used to. While lies, spin and deception can win a party an election, such tactics cannot win over the trust of voters. But as Ged Kearney and AOC have shown, major parties can easily overcome mistrust by preselecting candidates who are trusted not just by party insiders but also by the electorates they seek to represent.

Does a good candidate add 1 per cent to the party vote? Two per cent? Ten per cent? In 2016 the ALP candidate for Batman, David Feeney, received 35.3 per cent of the vote and, after preferences, narrowly defeated the Greens’ Alex Bhathal, who outpolled him with 36.2 per cent. In the 2019 election, the election in which there was a swing away from Labor of 1.4 per cent, Kearney delivered a swing of 11.7 per cent to deliver a primary vote of 46.8 per cent. The vote for the new Greens candidate, David Risstrom, collapsed by 15 per cent. Candidates matter.

But perhaps Jacqui Lambie provides the most interesting story about the relative role of money and candidate support in Australian politics. Lambie was elected to the Australian senate in 2013 on the back of Clive Palmer’s first, relatively successful, and relatively cheap, tilt at politics. Like most people elected to the senate on the Palmer United Party ticket in 2013, she quit the party to sit as an independent. While she was relatively unknown when she ran in 2013, Lambie used her time in the parliament to build her profile and, more importantly, a reputation as a straight shooter. In 2019, when, with virtually no funding, she faced off against the best-funded political campaign in Australian history, Lambie’s primary vote of 8.9 per cent was more than three times that of her United Australia Party rival.

Lambie’s return to the senate was built entirely on her credibility with Tasmanian voters and owed nothing to cheques from big donors. Money matters in Australian politics, but not as much as people think.

This article was first published in the print edition of The Saturday Paper on Jul 6, 2019 as "The maw of averages". Subscribe here.

Richard Denniss
is The Australia Institute’s chief economist.