Politics
Negative campaigns work, but not in the way we may think
The Coalition’s deceitful and damaging campaign against the Voice was highly effective. Over the course of the campaign support for the Voice fell from 65 percent at the start, to 40 percent at the actual vote.
By contrast, the Morrison government scare campaigns in the 2022 general election backfired spectacularly against the Liberal Party and their backers in Advance Australia. Independents and Greens either took seats from the Coalition or consolidated their positions, while the Labor vote fell.
These contrasting outcomes confirm research, published in the journal Political Communication: The fleeting allure of dark campaigns: backlash from negative and uncivil campaigning in the presence of (better) alternatives, by Philip Mendoza, Alessandro Nai and Linda Bos. They find that in a two-party contest a negative campaign, in some circumstances, may help the party mounting the campaign. But where there is at least one prominent alternative (independent or other party), that alternative candidate may benefit.
The Voice was a simple two-way contest. Where there was a prominent and politically attractive alternative to the offerings from the two traditional parties, both tended to lose, to the benefit of the alternative – the “teals” for the most part.
Dunkley by-election – a few things apart from the obvious
The obvious: Labor held the seat
Labor won, even though its TPP position has been pared from 56.3 percent at the 2022 election to 52.7 percent. That’s a swing of 3.6 percent, maybe a bit more than the normal by-election swing. The Liberal’s primary vote is up 6.6 percent and Labor’s is up 0.9 percent. A full set of figures can be found on Antony Green’s website.
In these figures there’s not much that points to any pro- or anti-government movement since the 2022 election.My rough assessment, based on primary votes, is that the “left” vote was unchanged at 51 percent, and the the “right” vote was unchanged at 47 percent.[1]
The not-so-obvious
The media has focused on the Labor-Liberal contest, as they have been conditioned to do for the last 80 years. In fact this by-election was closer to a binary contest than the 2022 election. In 2022 the Liberal + Labor vote was 72.7 percent, in this by-election it was 80.2 percent. Because it was brought on by the unexpected death of the sitting member, there was little time for independents or small parties to mobilize.
Journalists and the old political parties have been keen to draw inferences from the outcome. On the ABC Insiders the following morning four well-informed political observers – David Speers, Nikki Savva, Patricia Karvelas and Paul Sakkal (Age and SMH) – noted that the mood in the Liberal camp was upbeat because of the primary and TPP swings. Dan Tehan was almost jubilant about “a very good result for the Liberal Party”, announcing to the world “we’re back”.
Cooler heads would have pointed out some hard facts.
Dunkley, 50 km from Melbourne’s CBD, is just the sort of outer suburb that Liberal strategists hope to seize from Labor. Booth-by-booth outcomes, however, suggest that in the more working-class parts of the electorate there was no swing to the Liberal Party. The Liberals did enjoy a swing in the more prosperous parts, such as Mount Eliza, but would this have been the case had there been a strong independent standing on a “teal” type platform? (The one independent, Darren Bergwerf, increased his vote from 3.9 percent in 2022, to 4.8 percent in this by-election. Bergwerf, founder of the My Place network, is a long way ideologically from the “teals”.)
There could hardly be a worse time for a government to face a by-election than four months after the thirteenth interest rate rise, when real wages have been falling, when consumer prices have been rising, and when tax cuts are still four months away.
The Liberals’ 6.6 swing is largely, or perhaps entirely, explained by the absence of One Nation, and UAP, who won 7.9 percent in 2022. Applying 2022 preference flows I calculate that 5.3 percent of the Liberals’ vote came from their preferences.
The Liberals approached the campaign with a Duttonesque fear campaign about refugee boats and asylum seekers, supported by Advance Australia’s “put Labor last” campaign on the same issues. It is possible that the Liberal Party politicians and strategists cannot shake off a patronizing attitude to the working class, believing that the unschooled proletariat cannot see through scare campaigns, lies and bullshit. The ranks of the unschooled are thinning and ageing: demography is not friendly towards the Coalition’s political style.
It was possible for the Liberals to keep Dutton away from the campaign, which, as a by-election, was largely local. But in a full election the party leaders are prominent, whether the parties wish it or not.
As Nikki Savva said, the party is in a state of denial and delusion, a delusion that’s hard to break.
The Liberal Party is unshakeable in its delusion about its economic competence. It is possible that the Liberal Party has contributed to that delusion by the way it has defined economic management. That definition has almost always been in terms of the Commonwealth’s fiscal outcome, as if balancing the cash budget is the most important, or even the sole, aspect of economic management. They have come to believe their own spin.
The economy is actually about much more than the cosmetics of bookkeeping. Over nine years of Coalition government to 2022 almost all important economic indicators – real wages, productivity, the condition of public services, the adequacy of infrastructure, education outcomes – were going in the wrong direction. People may rate the Coalition positively in polls about fiscal bookkeeping, but on important economic issues the Coalition has not delivered.
Leading up to this by-election the Liberals made a lot of noise about the cost-of-living, but as Laura Tingle pointed out before the by-election, they offered no policies to alleviate people’s financial stress.
In attacking the government over the cost of living, the Coalition sends the message that the government cando something, in the short term, to relieve people’s fiscal stress. Therefore the alternative government should be able to suggest something. But the hard reality is that they cannot, because our cost of living problems are the result of decades in which governments, predominantly Coalition governments, have neglected structural reform and have underinvested in physical and institutional infrastructure, and in human capital – a consequence of the “small government” ideology.
Largely unnoticed was a 3.9 percent fall in the Green vote, from 10.3 percent in 2022 – a high result for an outer suburban electorate. Is this perhaps a reaction against the Greens’ Senate tactics of (rightly) identifying the government’s reforms as too timid, but forming alliances with the Coalition to block reforms altogether. Many people remember the Greens’ behaviour in 2009, when their intransigence in the Senate set back action on climate reform for 13 years, and led to the political ascendency of Tony Abbott, Scott Morrison and Peter Duttonin the Liberal Party. Or is it simply that they decided to run dead as some observers have suggested? If that is the reason, it’s a revelation about the cost of not campaigning and by implication the effectiveness of campaigning.
The outcome is probably good for the Labor Party, because they can read it as a vindication of their policies, and because it has not prompted a move to replace Dutton. But by the same token it is not good for Australia, because it delays the reconstruction of the Liberal Party, or its replacement with a centre-right party with policies that address national needs.
1. My rough classification was to include in the “left” primary votes for Labor, Australian Democrats, Greens and Victorian Socialists, in the “right” primary votes for Liberal, Libertarian, Liberal Democrats, One Nation, UAP, Australian Federation and independent Darren Bergwerf, and not to assign Animal Justice as left or right. ↩
The Albanese government grovels to the gambling lobby
Patricia Karvelas noted that Peta Murphy, whose death precipitated the Dunkley by-election, had been a passionate advocate for restriction of online gambling.
Murphy was Chair of the Parliamentary inquiry into online gambling, which reported last June: You win some, you lose more. Announcing its main recommendation, that online gambling should be banned within three years, Murphy said:
Gambling advertising and simulated gambling through video games, is grooming children and young people to gamble and encourages riskier behaviour. The torrent of advertising is inescapable. It is manipulating an impressionable and vulnerable audience to gamble online.
Labor politicians have been generous in their eulogies for Murphy, but have been even more generous to gambling lobbies.
Writing in Michael West Media – Murphy Report on gambling reform delayed by sports bet lobby – Rex Patrick and Philip Dorling report that Communications Minister Michelle Rowland has opened her door to gambling interests – parties that didn’t submit to the open process of the Parliamentary inquiry, and prefer the secretive process of ministerial lobbying.
Will social media keep providing news feeds?
Probably not, going by the announcement by Meta (Facebook’s parent company) that it intends to stop paying for Australian news content.
Peter Greste describes the consequences of this decision in his Conversation contribution: How will Meta’s refusal to pay for news affect Australian journalism – and our democracy?
Adversely, is his answer. He explains:
A key part of any successful democracy is a free media, capable of interrogating the powerful and holding governments to account. Even in a world overflowing with digital content, we recognise the need for good journalism, produced to ethical and professional standards, to help inform public debate and good policy-making.
He points out that the News Media Bargaining Code, negotiated by the Commonwealth government in 2021, was useful in providing a news feed to social media users, and in providing revenue – about $250 million a year – for Australian journalism. But he also points out the fragility of this arrangement in trying to have a commercial outfit made responsible for providing a public good, that public good being access to news. The hardheaded reality is that providing news has no benefit for social media companies, particularly when they have to pay for it, and its customers aren’t particularly interested:
Meta doesn’t need high-quality news, particularly if its users are more interested in sharing family photos than sober reporting on inflation rates. But collectively, our society does need it.
Greste also reminds us that the way we take in news has changed. The Bargaining Code was about making articles, originally designed for print media, accessible through social media, but people are now being informed (or misinformed) in different ways.
Above all we need to fund quality journalism – and quality journalism is expensive. “If we accept that news is a public good, not something we can treat as a product to be traded like soap, then we have to develop economic models that somehow get the public to pay for it” he writes.
In an interview on ABC Breakfast – Rod Sims “extremely disappointed” by Meta's 'arrogant behavior’ – Sims explains the problem not only in terms of what will be lost, but also what will come to dominate social media:
Journalism is a fundamental underpinning of our society. It’s not called the fourth estate for nothing. It’s a crucial part of how society functions: it holds powerful people to account; it’s an accurate journal of record; it’s a forum for ideas. And the less journalism you’ve got on social media the more you’ve got, potentially, of misinformation, disinformation. And of course the social media companies, the way they make money, is to keep you on the platform so they get your attention, they get your data, and they can advertise to you. And they have an awful temptation to bring material to you that outrages you, and therefore engages you and keeps you on the platform.
Perhaps that’s a little too generous to traditional media: the practice he describes seems to apply to the Murdoch tabloids as it once applied to The Truth. But at least their writers are identified and are bound by some attachment to a code of conduct. There is no such constraint on neo-Nazis, confused postmodernists, Advance Australia, Critical Race Theory missionaries, the Liberal Party, or sovereign citizens who put their drivel on social media.
Sims and others with an interest in media, including the ABC, Country Press and Crikey, also appear in an informative 5-minute segment on the ABC’s 7.30 program: Meta plans to stop paying for news content on Facebook.
Policymakers need to find ways not only to fund quality journalism, but also to encourage those who present it in ways that people will take notice of it.
For their part commercial media need to lift their game. The all-or-nothing model, allowing access only through a full subscription, is a relic from another era when people bought one daily paper produced for their own state (or half of the state in Tasmania’s case). This is particularly relevant for people living in Queensland, South Australia and Tasmania, served by daily Murdoch media. If they want to dip into articles in other media, such as The Age, past the occasional free article the only option is a full subscription, even if they have no interest in Victoria’s football, Victoria’s murders, Victoria’s roadworks and so on. We are accustomed to making small payments, debited from pre-loaded accounts, for toll roads, public transport and services such as Skype, but the marketing staff in newspapers are too unimaginative to develop pay-per-view models for quality journalism. That’s a textbook case of deadweight loss, where the public, the companies and ethical journalists lose out.
Artificial intelligence
A couple of readers have suggested I should occasionally include links on the policy implications of artificial intelligence. Because it’s an area with thousands of views, and it’s a long way from my specialization, I decided to ask Chat GPT, and to give a link to their their ten-point response.
That response might earn a first-year student a “pass”, or “resubmit” grade, with comments such as “Major issues about disinformation and existential risks mentioned, but no critical analysis, no identification of sources, and no weighting according to competence and possible interests of those sources. Conclusion is insipid”.
In this example lies one major problem with AI – its (apparent) neutrality on this particular platform, Chat GBT. But that’s its most innocent use.
One reader has sent a link to this ten-minute video of AI text-to video examples. The video clips could be called amusing, cute, or creative, and not deliberately misleading. But we can imagine how the techniques illustrated in this video could be used to mislead.
That same video has put the wind up film and TV producer Tyler Perry, who sees the potential for this technology to destroy jobs in the entertainment industry.
That’s not only about the usual issue of technology displacing employment. It’s also about dulling the distinction between fiction and reality, even in the absence of evil intention.
To go just one step further, we can imagine a clip of a large crowd of unmolested demonstrators in Moscow, freely campaigning for the opposition. Or a clip of dinosaurs and humans co-existing in a children’s religious “science” education program.
And that’s before we imagine more sophisticated applications involving manipulation of people’s images. The porn industry is already doing this, and it won’t be long before there are convincing clips of public figures using AI-created speeches – Volodymyr Zelenskyy, giving a Hitler-style speech to a crowd of swastika-bearing followers, Taylor Swift endorsing Donald Trump, Ross Garnaut extolling the benefits of fossil fuels.
They’re the scary aspects of AI. On the other side Nick Gruen has brought to our attention a talk by Andrew Leigh – Using artificial intelligence for economic research: an agricultural odyssey – demonstrating that AI can do a lot of the slog work in agricultural research – an area where there is heaps of data and where patterns are hard to detect.
And of course much AI goes unnoticed – from the way our photography can be edited through to city-wide optimization of traffic flows.