Why the Future Cannot Be Predicted

Abstract

In recent years, scholars of public opinion and democracy have noted two disturbing trends. First, across many countries satisfaction with democracy's performance has been declining. Second, while support for democracy as a system of government remains overwhelmingly high compared to any alternative, levels of democratic support have steadily, if rather gradually, slipped (Figure 1). These findings have sparked debate, including contentions over their possible relation to other recent trends such as the rise of populism, authoritarianism, and democratic backsliding. 1 The findings are also what led us, five years ago, to float the notion of democratic deconsolidation-the now fairly uncontroversial idea that, just as democracies can consolidate when elites and the public embrace liberal-democratic norms, so too can democracies "deconsolidate" once such norms begin to fray. 2 In the April 2021 issue of the Journal of Democracy, Christian Welzel argues that scholars of democracy and public opinion should look not at commonly fielded questions that ask citizens how they feel about the performance and value of democratic institutions in their countries, but should instead focus on survey items that probe attitudes about such topics as abortion, divorce, women's rights, and approaches to parenting. 3 If Welzel is correct, then we can all stop worrying about such unsettling

What Satisfaction with Democracy? A Global Analysis of “Satisfaction with Democracy” Measures

“…Seminal and widely cited texts in political science utilize the measure (Crozier, Huntington, and Watanuki 1975; Dalton 2004; Newton 2014; Norris 2011), and more recent book-length treatments have applied sophisticated quantitative methods to increasingly large amounts of data (Ferrin and Kriesi 2016; Martini and Quaranta 2020; van Ham et al 2017; Vowles and Xezonakis 2016; see also Foa et al 2020). However, most of these studies rely on a single survey project; to our knowledge, only two of these use data from multiple survey projects (Foa et al 2020; van Ham et al 2017). Although there is less “ambiguity and contradiction” now than when Canache, Mondak, and Seligson (2001, 510) provided their analysis of the meaning of the SWD measure, there is still confusion over the different trends and levels reported between different survey projects, regions, and time periods (see also Linde and Ekman 2003).…”

“…While they do account for sampling bias resulting from differential country inclusion, there is no indication that they account for bias resulting from differential survey inclusion in each year. This raises concern because many of the surveys they include in their analysis are only included in a few recent years and not equally across all countries (see Foa et al 2020, 6), and they do not examine trends separated by survey source. Therefore, their substantive results might be influenced by differential survey inclusion and influenced by the comparability and validity concerns we discuss in this paper.…”

“…Notions of a “crisis of democracy” in established democracies have been prevalent since at least 1975 (Crozier, Huntington, and Watanuki 1975; Dryzek et al 2019; Ercan and Gagnon 2014; Merkel 2014), while others have disputed any clear or uniform trend of decline in political support (Hibbing and Theiss-Morse 2002; Newton 2014; Norris 2011). Norris (2011) argues for trendless fluctuations and persistent country differences, while van Ham et al (2017) and Martini and Quaranta (2020) used data from the EB to conclude that SWD had in fact been increasing across established democracies, an opposite conclusion to the Foa et al (2020) report discussed.…”

“…Previous methodological literature focusing on the conceptual aspects of the measure has usually relied on a relatively limited data range. Those that are indeed interested in trends have often relied on data from only one or two survey projects, with the exception of Foa et al (2020) noted above, which does not take into account potential issues with differential survey source inclusion over time. Here, we aim to analyze the consistency of SWD measures between different data sources, while also shedding a light on the debate about trends in SWD and providing a guide for caution in future research utilizing the measure.…”

Midcentury Modern: The Emergence of Stakeholders in Democratic Practice

“…McCaffrie and Akram 2014; Urbinati 2019, 18). Such concerns are not unfounded: recent surveys suggest that popular dissatisfaction with democracy is particularly acute in the United States and generally rising across the globe (Foa et al 2020, 9, 19–20). Defenders of democracy often call for a renewal of participation, but such proposals for democratic engagement and participation miss a curious puzzle: the last few decades have seen increased opportunities for public participation in diverse domestic and international institutions (Rosenbluth and Shapiro 2018, 3).…”