How can we find out whether people are really turning against democracy?

Democracy is in decline – or so a growing consensus suggests. Paul Schuler sets out the evidence for claims that people are turning to autocratic alternatives, and asks whether they necessarily show a loss of faith in democracy. He proposes some alternative measures that could establish whether people are genuinely willing to trade freedom for a ‘solution’ to anxieties about immigration, inequality and globalisation.

donald trump

Donald Trump at CPAC 2011 in Washington, DC. Photo: Gage Skidmore via a CC-BY-NC SA 2.0 licence

Is global democracy in decline? The amount of scholarly firepower trained on the question suggests it is. The Social Science Research Council has commissioned a series of essays on the “Anxieties of Democracy.” In the last two years the Journal of Democracy has published a series of articles on this very subject, with most agreeing that the threat is real.

Unfortunately, despite evidence from specific elections that seems to justify the concern, current research designs and survey indicators do not seem able to describe or predict the outcomes generating it. What is the evidence for a democratic decline and its drawbacks? And are there alternative measurement strategies?

Evidence for a global decline in democracy comes in three types. The first form of evidence comes from cross-national indicators (typically Freedom House), showing that the number of free countries is either not growing or declining. The second type comes from election outcomes indicating an increase in support for leaders that appear less committed to democratic norms. The third type focuses on individual level survey data, typically from the World Values Survey. Of these three forms of evidence, changes in the cross-national indices seem the weakest. Although there has been some levelling off of Freedom House Scores, the decline does not yet constitute a “reverse wave.” Furthermore, as Levitsky and Way point out, what decline there is may have more to do with overestimating the true levels of democracy in the 1990s than any actual reduction in democracy.[1]

The more compelling evidence for the democratic decline comes from global election returns and the survey data. This evidence implies that while democratic deficit may not yet be reflected in the democracy indicators, the storm clouds are gathering. In terms of electoral returns, the success of parties and leaders advocating nationalist sentiments (some of whom have won) in Europe and the US has raised concerns that these appeals could be used to trample on liberal freedoms, particularly for minorities. The problem extends to Asia, where Prabowo Subianto, a former military official under deposed autocrat Suharto came close to winning the presidency despite (or because of) his appeals to roll back liberal democracy. In the Philippines, popular president Rodrigo Duterte has advocated vigilante justice, particularly against drug dealers and drug users.

Despite these concerning outcomes, we still lack the theory or evidence to link these disparate events. What makes these leaders or parties anti-democratic? Is it their willingness to trample on the civil liberties of minorities? Is it their willingness to undermine democratic institutions? The former is a threat to liberal values, while the second a greater threat to procedural democracy. In Europe (and in the US), the greater concern seems to be over anti-immigration, anti-globalisation nationalism. It is less clear that this translates outside of Europe to the Philippines, Indonesia, and Latin America, where the concerns centre on centralising power in the executive and dispensing with the rule of law. Both are obviously concerning, but the root causes and solutions are likely to be different.

The importance of diagnosing the nature of these anti-democratic appeals becomes important when we consider evidence for the democratic decline emerging from survey responses in the World Values Survey (WVS). Foa and Munck argue that democracy is under threat based on WVS data showing an increase in dissatisfaction with democracy and support for autocratic alternatives. Although the number respondents supporting autocratic alternatives remain in the minority, the numbers are increasing – particularly among younger respondents in the US and, to a lesser extent, Europe. This seems to suggest a growing rejection of democratic norms and support for autocratic alternatives, which should implicate both liberal norms and democratic institutions.

On the face of it, this evidence seems compelling. However, an easy interpretation is complicated when we consider that groups expressing dissatisfaction with democracy in WVS differ from those voting for the supposedly autocratic candidates. While the younger voters tend to exhibit less support for democracy in the WVS it is middle-aged and older voters that are more likely to support right wing candidates.

There are a number of reasons why this linkage might not be so easy. First, based on the “critical citizens” framework, it could be that dissatisfaction with democracy does not necessarily mean less support for democracy. Second, it could be that questions using the “d” word may not accurately measure support for the democratic values scholars have in mind when they construct the questionnaires.[2] In particular, confirmation bias or different conceptualisations of the term complicate interpretation of answers to these questions. As a partial solution, some surveys have attempted to ask about whether the respondent supports displacing parliament for a strongman, or allowing the military to rule. However, it is not clear whether or not these questions in isolation can capture support for autocracy or opposition to democracy. Indeed, in an abstract sense, most citizens would prefer not to have the military rule. However, this is not necessarily what we care about. The question is whether people are more willing to trade off democratic norms in the pursuit of certain goals.

Better ways of measuring people’s commitment to democracy

To get a better grasp on the nature of the problem, I have a few suggestions:

First, in addressing the potential linkages between the different campaigns, it would be useful to conduct text analysis of campaign speeches across countries and across time to assess whether or not there are similarities between anti-democratic candidates and parties across country contexts. Are the messages the same? Does we see a qualitative and quantitative shift from past threats to democracy?

Second, more work should be done to link the questions asked in surveys like the WVS directly to voting behaviour. Where we can identify candidates that appeal to anti-democratic tendencies, do we see voters expressing less confidence in democratic institutions voting for them?

Third, we must devise measures that better assess how citizens trade off support for democratic norms against their other substantive objectives. One potential avenue could be a greater use of contingent evaluation questions, which are most commonly used in market research. The most basic example is how much would consumers be willing to pay for increases in quality. Because simply asking whether consumers want high quality goods (“Yes!”) or lower prices (“Of course”) is unlikely to be interesting, these questions force respondents to identify the price they would pay for increases in quality.

In terms of support for democracy, one could ask respondents the degree to which they would trade support for democratic norms for specific substantive goals. How much are respondents willing to “pay” in democracy for a “win” on the range of issues (immigration, globalisation, inequality, environment). Would citizens support suspension of that country’s institutions to have a leader unilaterally impose a solution to one of these issues?


[1] Levitsky, Steven and Lucan Way. 2015. “The Myth of the Democratic Recession.” Journal of Democracy.

[2] Chu, Yun-Han and Min-hua Huang. 2010. “Solving an Asian Puzzle.” Journal of Democracy.

This post represents the views of the author and not those of Democratic Audit.

paul schulerPaul Schuler is an Assistant Professor at the University of Arizona School of Government and Public Policy. His research centres on authoritarian politics and democratic transitions. His regional focus is on Southeast Asia, with a particular specialisation on Vietnam. Previous work by Dr Schuler has appeared in the American Political Science Review, Legislative Studies Quarterly, and the Journal of East Asian Studies. 

Similar Posts

Posted in: Populism, Uncategorized