What is democratic backsliding?

Democratic backsliding is when a democratic country shows signs of becoming autocratic or authoritarian. Backsliding can occur when a democracy has just a foothold (e.g. Poland in the early 1990s) or is firmly established as a democracy (the USA).

How do we know when democratic backsliding occurs?

Social scientists typically use democracy measures, such as Freedom House, or Varieties of Democracy (V-Dem), or the Global State of Democracy, as a benchmark. First, they measure democracy in one year. That is their benchmark. Then, to measure change, in a subsequent year, they measure democracy again. A country that has a lower score from year to year may be backsliding.

However, democracy measures can have problems. A major problem is that they may not pick up smaller, more subtle signs of backsliding.

Enter Roberto Foa & Yascha Mounk. Their famous 2016 article, The Danger of Deconsolidation, used the World Values Survey, a cross-national survey dataset of many countries around the globe, to understand who supports democracy. They argued that major democracy measures do a passable job, but we also need to understand, from the ground level, changes in mass support for democracy.

In this post, we examine Foa & Mounk’s argument and some of their critics.

Related to this article

• Political Participation and Democracy
• Political Equality as Measured by Varieties of Democracy (V-Dem)
• Democracy and Economic Inequality
• The Many Definitions of Political Inequality

History of Democratic Backsliding Studies

The concept of “democratic backsliding” is also called “democratic deconsolidation.” An “established” democracy is consolidated. When it changes to authoritarianism, it has “deconsolidated.”

Democratic consolidation was a popular term in the 1980s and 1990s, around the time that the US had a policy of democracy promotion around the world. Around that time there was a proliferation of quantitative democracy measures.

Foa and Mounk in 2016 revived the term by trying to sound the alarm on “deconsolidation.” Because Foa and Mounk did not properly acknowledge the long history of democratic consolidation studies from the 1990s, they obscured those early studies’ original purpose, which was to sound the alarm on possible deconsolidation.

Foa and Mounk’s critics missed the point. We should be looking for the small and troubling signs of democratic backsliding. Foa and Mounk’s (F&M) 2016 article outlasts their critics because their fundamental point was correct, even if their measures of democratic backsliding had some flaws.

Democratic Backsliding is about Transition

Consolidation is mainly seen as a process from transition democracies to consolidated democracies. The concern has always been the survival of democratic regimes, and thus intrinsically about democratic backsliding. The emphasis of the 1990s literature seemed to be on how transition societies – especially Latin America and Eastern Europe – could solidify their democratic gains into long term stability. 

But Consolidated/Consolidation have always been fuzzy concepts. The various definitions can be compared and contrasted, but in the end (it’s sometimes called, “democratic decay”), there has been no singular definition of what a consolidated democracy looks like or what the process of consolidation entails. There are some similarities across authors’ arguments. F&M’s definition is a good place to start, but in the end, they do not offer enough specifics to identify a consolidated from a transitional democracy.

The literature has tendrils in many topics, such as democracy, democratization, states and regimes, transitions and development, political behavior (voting especially), and democratic values, but also civil society, bureaucracy, and economic development.

In the end, the core idea is that democracy is under threat of backsliding.

Foa and Mounk’s The Danger of Deconsolidation

Foa & Mounk seek to warn us that we may be unjustifiably complacent about the well-being of consolidated democracies. We have not anticipated other extreme events (like the collapse of the USSR) and we may be in the midst of one now. 

The authors note that, in North America and Western Europe, trust in institutions (such as parliament and the judicial system), party membership, and voter turnout has declined, and party identification has weakened. Voters are turning to anti-establishment parties, fueling a rise in populism. In these stable regions of the world, democracy seems to be in trouble.

Critics of the “decline of democracy” approach (Inglehart, Wezel, Norris, Dalton) argue that while support for particular governments regularly declines (what they call government legitimacy), support for democracy itself (what they call regime legitimacy) remains robust. The people know that democracy allows them these expressions of discontent and thus support the regime, but not the government.

F&M feel that that the critics argument is optimistic. They seek to challenge that view.

F&M use waves 3 through 6 of the World Values Surveys (1995–2014). With these data, they attempt to measure four types of regime legitimacy:

1. Support for the whole system
2. Support for “key” institutions of liberal democracy, e.g. civil rights
3. Willingness to advance political causes
4. Openness to authoritarian regimes such as military rule.

In their presentation style, they look at extreme values. The point of the article is to provoke and hunt for any sign, no matter how small, of deconsolidation.

1. Support for the whole system

In Figure 1, they measure support for the whole system with the “Percent of respondents rating it ‘essential’ (a rating of 10 on a 10-point scale) to ‘live in a country that is governed democratically’” (p. 7). They compare the US with “Europe.” The X axis is birth cohort by decade (1930s to 1980s) and the Y axis is percent that rated democracy as essential. Both the US and Europe show a negative relationship. The older cohorts (1930s to 1950s) still support democracy at above 50 percent. The younger cohorts (1960s to 1980s) are at 50 percent or less.

In Figure 2, their second measure of regime support is with “Percent responding that ‘having a democratic political system’ is a ‘bad’ or ‘very bad’ way to ‘run this country,’ by age group.” They compare age groups in the US and Europe. Comparing age groups in the US as of 2011, the authors find a range of ca. 12 percent to nearly 25 percent, with older age groups evincing lower percentages. They find something similar in Europe, but the range is very small (from 6 percent to ca. 13 percent).

In sum, older people are more likely to support the regime than younger people.

2. Support for “key” institutions of liberal democracy, e.g. civil rights and 3. Willingness to advance political causes

It is possible that people can support democracy but not support its institutions or politically participate. Here, they don’t have graphs and don’t offer many numbers (but see Fig. 3).

They find that millennials support the idea that it is absolutely essential in a democracy for civil rights to protect liberty less (32 percent) than those born in the interwar and immediate post war environments (41 percent). The spread for Europe is much smaller (39 to 45 percent).  They also find that, in the US, 14 percent of baby boomers argue that it is unimportant in a democracy that people “choose their leaders in free elections” as compared to millennials, 26 percent make that argument. In Europe, the spread is smaller and ranges from 9 to 13 percent. They looked at other regions of the world and did not find the same result.

Foa and Mounk claim that there is a widening gap between age groups in “political apathy.” Older cohorts are more likely to be interested in politics and to engage in political participation (both conventional/institutional and unconventional/non-institutional).

In sum, older cohorts are more likely to support the key institutions of democracy than post-baby boomer cohorts.

4. Openness to authoritarian regimes such as military rule

Here, they look for support among Americans and Europeans for military rule that they consider as an anti-democratic idea. Unlike the previous sections, in this section they combine income with age. 

First, age: Overall, there is a trend in Americans who believe that it would be a good or very good thing if the army ruled the country (from 1 in 16 to 1 in 6). They note a predictable age gap in this attitude. “In Europe, the generation gap is somewhat less stark but equally clear, with 53 percent of older Europeans and only 36 percent of millennials strongly rejecting the notion that a government’s incompetence can justify having the army “take over”” (13).

Then, income. The authors looked at income groups and conclude that “whereas two decades ago affluent citizens were much more likely than people of lower income groups to defend democratic institutions, the wealthy are now moderately more likely than others to favor a strong leader who can ignore democratic institutions” (13).

And then, the combination of age and income. “In Europe in 1995, 6 percent of high-income earners born since 1970 favored the possibility of “army rule”; today, 17 percent of young upper-income Europeans favor it” (14).

In sum, they conclude that the affluent, the young, and the young and affluent are more likely to support military rule than other age and income groups.

“Is Democracy Deconsolidating?”

They ask the big question of whether all of this adds up to democratic deconsolidation. The authors present the finding of Przeworski and Limongi that “no consolidated democracy with a GDP per capita of over $6,000 in 1985 international prices has ever collapsed.” 

The authors claim that this finding has blinded further research in the idea that consolidated democracies can deconsolidate. In this article they address whether data can tell us if stable, wealthy, and consolidated democracies can become unstable and deconsolidated. 

How do we know if a democracy is consolidated? The authors quote Linz and Stefan: democracies are consolidated when they are the “only game in town.”

But the authors disagree with the premise, as they question how we would know if democracy is the only game in town. At the end of the article, Foa and Mounk offer their indicators of consolidated democracy:

“In our view, the degree to which a democracy is consolidated depends on three key characteristics: the degree of popular support for democracy as a system of government; the degree to which antisystem parties and movements are weak or nonexistent; and the degree to which the democratic rules are accepted.” (15)

In this article, they looked at “popular support for democracy,” but did not look directly at the degree to which antisystem parties are weak, or directly at the acceptance of democratic rules other than support for civil rights.

The authors note the rise of Trump, the rise of right wing populist parties, and the decline in approval of mainstream and long-established politicians as indicators of a challenge to democratic consolidation. As they summarize:

“Citizens of democracies are less and less content with their institutions; they are more and more willing to jettison institutions and norms that have traditionally been regarded as central components of democracy; and they are increasingly attracted to alternative regime forms.” (16)

Democracies that begin to deconsolidate may not fail, and democracy may not fall out of favor they argue. But, the signs of deconsolidation are apparent, they believe. 

Critics of Foa and Mounk Miss the Point

Critique by Ronald Inglehart

Ronald Inglehart challenged the thesis of Foa and Mounk in a 2016 reply called “How Much Should We Worry?”, also published in the Journal of Democracy.

Inglehart argued that the strongest effects of democratic backsliding, as measured by Foa and Mounk, are in the US, and thus F&M’s argument is mostly about America. Inglehart blames political dysfunction, growing economic inequality, and growing political inequality.

Inglehart then adds “value change.” The societies that F&M examine are undergoing a shift from materialism to post-materialism. It is a movement from insecurity to security. Secure people are more tolerant and tend to support democracy. Yet, within these societies, there are people who face an “existential insecurity” and the economic crises has exacerbated this sense. Existential insecurity means greater support for authoritarianism, xenophobia, and a breakdown of norms.

The young are particularly vulnerable, Inglehart argues: “Existential security has been declining for most of the population—especially the young, who face high levels of unemployment, even among those with university or postgraduate educations”.” (21).

In the long-run, modernization will win out. Why? Economic development leads to democracy because it creates the conditions in which democracy can flourish – economic security, an educated workforce, and a rise in self-expressive values.

When one argues that modernization leads to democracy, it is a classic way of theorizing: looking back in time, tying together trends, calling the trend something – modernization, in this case – and then declaring it a theory that would predict such a thing.

Critiques of Data and Methods

The other critics of Foa & Mounk’s 2016 The Dangers of Deconsolidation tend to attack the data and methods, writing that the signs are too small to matter or can be erased if one uses different measures or a different interpretation of the results.

For example, Alexander and Welzel argue that

“Foa and Mounk heavily overstate the age differences in democratic support. Second, the obvious age pattern in indicators of political disaffection has little to do with generations; it is instead a lifecycle effect: younger people showed stronger signs of disaffection already in earlier decades, but this age pattern is not linked to a uniform temporal trend towards increasing disaffection in the electorates of mature democracies…

Alexander and Welzel are right in that a core problem of democratic backsliding is political inequality:

“The source of the problem is certainly not the younger generation and its alleged loss of support for democracy. Instead, it is the growing marginalization of the lower social classes, their resulting ideological divergence from the increasingly progressive mainstream and the failure of the established parties, as well as the media, to adequately address the legitimate concerns of the “left behinds.””

Pippa Norris argues that, although backsliding has occurred in some countries, it has not done so in the West.

“Culturally, when more systematic survey data is examined across a broader range of more than two-dozen Western democracies and over a longer time period, in fact the claims by Foa and Mounk fail to prove consistently reliable and robust. The generational gaps presented by the authors are exaggerated both by cherry-picking cases and by the visual presentation and treatment of the survey data. Far from a uniform ‘European’ pattern, countries vary widely in public perception of democratic performance and persistent contrasts are observable. The data also suggests a persistent life-cycle effect.”

Erik Voeten argues that there simply has been no change.

“Millennials are not very different in their views of political systems than were young people in the mid-1990s. The evidence suggests that millennials in the U.S. are somewhat more skeptical of democracy than people of similar ages were twenty years ago. Nevertheless this evidence comes from one survey. Moreover, when we look at confidence in actual democratic institutions, then the opposite pattern emerges: older generations have lost faith in U.S. Congress and the Executive to a greater extent than millennials.

The take-away is not that there is no threat to consolidated democracies but rather that this does not come from abstract procedural preferences among (some part of) the populace for alternative regime types.”

However, these critics miss the point of Foa and Mounk: there are small and troubling signs of deconsolidation. The signs may be small. But they are troubling. Social scientists tend to miss major historical happenings and then jump to explanations of them after they occur. Foa and Mounk warned us of this in the first paragraphs of their article.

Foa and Mounk argue that, to prevent democratic backsliding, we need to pay attention to the small changes.

Further Reading

Acemoglu, Daron, and James A. Robinson. 2005. Economic Origins of Dictatorship and Democracy. New York: Cambridge University Press.

Alexander & Welzel. 2017. “The Myth of Deconsolidation: Rising Liberalism and the Populist Reaction” Journal of Democracy.

Diamond, Larry. 1999. Developing Democracy: Toward Consolidation. Baltimore: Johns Hopkins University Press.

Kwak, Joonghyun, Irina Tomescu-Dubrow, Kazimierz M. Slomczynski, and Joshua K. Dubrow. 2020. “Youth, Institutional Trust, and Democratic Backsliding.” American Behavioral Scientist 64, no. 9: 1366-1390.

Linz, Juan J., and Alfred Stepan. 1996. Problems of democratic transition and consolidation: Southern Europe, South America, and postcommunist Europe. Baltimore: Johns Hopkins University Press.

Przeworski, Adam, Michael E. Alvarez, Jose Antonio Cheibub, and Fernando Limongi. 2000. Democracy and Development: Political Institutions and Well-Being in the World, 1950–1990. New York: Cambridge University Press.

Copyright Joshua Dubrow Politicalinequality.org 2022

1. What is democratic backsliding?
2. How do we know when democratic backsliding occurs?
   1. Related to this article
3. History of Democratic Backsliding Studies
   1. Democratic Backsliding is about Transition
      1. In the end, the core idea is that democracy is under threat of backsliding.
4. Foa and Mounk’s The Danger of Deconsolidation
   1. 1. Support for the whole system
      1. In sum, older people are more likely to support the regime than younger people.
   2. 2. Support for “key” institutions of liberal democracy, e.g. civil rights and 3. Willingness to advance political causes
      1. In sum, older cohorts are more likely to support the key institutions of democracy than post-baby boomer cohorts.
   3. 4. Openness to authoritarian regimes such as military rule
      1. In sum, they conclude that the affluent, the young, and the young and affluent are more likely to support military rule than other age and income groups.
5. “Is Democracy Deconsolidating?”
6. Critics of Foa and Mounk Miss the Point
   1. Critique by Ronald Inglehart
   2. Critiques of Data and Methods
      1. Foa and Mounk argue that, to prevent democratic backsliding, we need to pay attention to the small changes.