Inside Story

Which crisis of trust?

Are concerns about Australians’ faith in politics and democracy being exaggerated by poorly presented research?

James Frost 18 July 2019 1707 words

Trustworthy? Senators and members assemble in the Senate at the start of the forty-sixth parliament on 2 July. Sam Mooy/AAP Image

We have lost faith in our leaders… Members of Parliament and Civil Servants… live like Lords — while we have to cadge for food — to Hell with you all.
— Anonymous, Collingwood

Australian politics is in the midst of a crisis of trust. Australians are rapidly losing faith in politicians; young voters are increasingly disillusioned with politics; older voters are sick and tired of the endless cycle of new prime ministers. This has been the growing consensus since academics began routinely conducting public attitude surveys in the 1980s.

And the situation seems to be getting worse. Surveys by the Scanlon Foundation and the Democracy 2025 project — a collaboration between the University of Canberra’s Institute for Governance and Policy Analysis and the Museum of Australian Democracy — show steep falls in political trust. So does the longest-running, best-known and most influential large-scale survey, the Australian Election Study, or AES, which began in 1987.

The AES’s much-publicised overview report, Trends in Australian Political Opinion, released in late 2016, compares results for 152 survey questions asked over three decades or more, using both its own data and findings from surveys conducted by political scientist Don Aitkin in 1967, 1969 and 1979. Of the graphs in the report, two have propelled the “trust crisis” narrative like no others. In one, 40 per cent of survey respondents say they are “not satisfied with democracy”; in the other, just 26 per cent agree that “people in government can be trusted.” The Sydney Morning Herald’s headline — “The Fundamental Operating Model of Australian Politics Is Breaking Down” — was typical of the media’s response to these findings.

But that headline made me wonder why, if political trust is so low, Australia isn’t on the verge of revolution. The AES findings suggest a real feeling of anger with politics, politicians and political institutions, yet they don’t seem to fit with day-to-day reality. Curious, I started looking more closely at opinions expressed about politicians and democracy in mainstream newspapers.

Here, too, the evidence wasn’t reassuring. According to people who write letters to the editor, politicians “always think of themselves” and invariably break their promises. They are “loud-mouthed careerists” with nothing to offer but “words, words, words.” And, of course, they’re “out of touch” with the concerns of the average citizen.

Running parallel were complaints aimed at voters by editors, academics and MPs themselves. An editorial criticised Australian voters because they “don’t know one candidate from another” and are mostly “apathetic” about politics in general. Certain ex-politicians and academics viewed voters as self-interested, disengaged political novices who, through “a system of crooked bargaining,” sell their votes to the highest bidder. It’s a discouraging scene.

But I should come clean here. Every quote in those last two paragraphs, along with the quote at the top of this article, appeared in Australian regional and city newspapers and magazines in the early 1920s, the 1930s or the late 1940s — periods of social instability, political unrest, economic recession and even war. Yet they seem curiously familiar.

Jump forward a generation, to the late 1960s and early 1970s, and public trust in politics seems to have been even lower. “Australians are fed up with politics,” reported the Australian in April 1974, adding that widespread disillusionment with elections was part of “a general malaise.” According to one young would-be MP, “The Australian government ignores public opinion… it ignores voters… this country is run on every level by men who regard people as puppets, nothing more.”

Some young Australians were so disgruntled they formed a “people’s parliament” where “all the people ignored and neglected by the [government]… will have a chance to say their piece.” Compare this to young Australians today, who, according to one interpretation of the AES’s data, “display a greater willingness to flirt with authoritarianism.” These are not exactly the “revolutionaries” that some of their parents were.

I’m not the first to point this out. In 2013, worried that the “trust crisis” narrative was dominating academic and public debate, historian Jackie Dickenson wrote Trust Me: Australians and Their Politicians, from where many of the quotes above are drawn. Her intention was to “challenge the assumption that political trust today is at an all-time low,” because an unquestioning acceptance of that belief “feeds cynicism and apathy, and threatens the engagement of voters in politics.”

Dickenson’s work is part of a growing literature critical of the accuracy and consequences of the public attitude surveys that journalist draw on when they’re writing about trust. In a critique published thirty-one years ago, social scientist Diego Gambetta argued that “trust” — just like “freedom,” “knowledge” and “justice” — is an “elusive” concept with no fixed meaning. More recently, another scholar, Guido Möllering, identified “three perspectives” on trust, all of which contain further different ways of understanding the concept.

If “trust” can mean various things to one person, how can we be sure that social scientists are measuring the same idea when they survey 2000 respondents? And if we can’t be sure about definitions, how likely is it that surveys are assembling coherent sets of data? Yet the findings are rarely accompanied by caveats like these, so news of the crisis spreads unimpeded via the media and think-tanks through the public sphere.

Epistemological challenges aside, more straightforward problems are created when data are matched up from different sources. The two AES graphs I mentioned above are good examples: they draw on Don Aitkin’s 1969 and 1979 surveys, which were conducted as one-on-one interviews, and on AES data, which was gathered using paper questionnaires.

On the first theme covered by these graphs, Aitkin’s 1969 and 1979 surveys asked, “On the whole, how do you feel about the state of government and politics in Australia?” and provided four permissible answers: “very satisfied,” “fairly satisfied,” “not satisfied” and “don’t know.” The AES’s question, which it began asking in 1996, was significantly different: “On the whole, are you satisfied, fairly satisfied, not very satisfied or not at all satisfied with the way democracy works in Australia?”

Despite these differences, the AES’s 2016 report combines the two sets of data — its own and Aitkin’s — into a single graph under the headline, “Satisfaction with Democracy,” charting “yes” and “no” responses from 1969 to 2016. But Aitkin’s question didn’t make any reference to “democracy,” and included the word “feel,” which may have subtly influenced the way people answered. The differences between Aitkin’s question and the AES’s are referred to only in the endnotes to the report.

The AES’s “Trust in Government” graph also uses Aitkin’s 1969 and 1979 surveys as its starting point. Aitkin’s original question, “In general, do you feel that the people in government are too often interested in looking after themselves, or do you feel that they can be trusted to do the right thing nearly all the time?” had three possible answers: “Can be trusted,” “Look after themselves” and “Don’t know.” In 1993 the AES began asking a modified version of Aitkin’s question, “In general, do you feel that the people in government are too often interested in looking after themselves, or do you feel that they can be trusted to do the right thing nearly all the time?” with four possible answers: “Usually look after themselves,” “Sometimes look after themselves,” “Sometimes can be trusted to do the right thing” and “Usually can be trusted to do the right thing.” Again, the differences create obvious continuity problems, and once more a short explanation in the report’s notes doesn’t seem adequate.

Who can truthfully claim that one person’s feeling of satisfaction about “government and politics” is the same as another person’s professed support for “democracy”? It’s hard enough to identify a shared definition of what democracy means today, let alone confidently claim that 2000 respondents had the same “democracy” in mind when they were asked about government and politics fifty years ago.

All this suggests that the “trust crisis” narrative might be at least partly due to how the results of public attitude surveys have been presented. That reporting may then create a feedback loop: after years of hearing the crisis mentioned by media reports based on values surveys, individuals are likely to be more suspicious of politics in all its forms, and answer surveys accordingly.

Don Aitkin was aware of the importance of putting his findings in a proper historical context. In his highly regarded book Stability and Change in Australian Politics (1982), which compares his 1969 and 1979 surveys, he is at pains to explain that “survey evidence would have had little meaning without the prior work of many historians, political scientists and sociologists.” Aitkin knew that survey work required more than simply acquiring data and presenting it to the world.

Public expectations can be shaped by seemingly factual information about the state of politics repeated over and over by the media — just as the Coalition’s recent victory was only a surprise because of public expectations created by polls predicting a solid Labor win. Pollsters around the country have had their profession tarnished by methodological errors and a failure to make clear the limitations of their findings. Indeed, two of Australia’s leading news outlets, the Age and the Sydney Morning Herald, have decided to stop reporting polls for the foreseeable future.

The real value of surveys like the AES risks being undermined by the fact that they are often treated as the only way of gathering useful information about contemporary public attitudes. An overreliance on any single theory, analytical method or data-gathering technique only impoverishes the social sciences.

Perhaps with these concerns in mind, Democracy 2025’s first report, published in 2018, combined online questionnaires with twenty focus groups to measure public attitudes, and also explained its methods of analysis in detail. But its analysis, too, has led to alarmist headlines including “Australians No Longer Trust Their Democracy,” again highlighting the importance of historical contextualisation.

A final word on public trust in politics. We rarely acknowledge that liberal democracy requires a certain degree of distrust. Too much is a bad thing, of course, but we’re not at crisis point yet. Perhaps trust in politics is fluctuating as it always does, and is at a low ebb because of recent turbulence? If history is any guide, Australians will eventually return to merely disliking their politicians. We can only hope the “trust crisis” narrative has not diminished or fundamentally damaged the chances of this correction taking place. •