The territorial and qualitative expansion of democracy reached its global highwater mark in 2005, and since then the world has experienced an ebb of democracy, or so claimed Larry Diamond in his 2015 article published in the famous “Is Democracy in Decline?” issue of Journal of Democracy. Although the thesis of the democratic ebb has been intensely debated, in the last couple of years there appears to be a broad consensus among democracy-measuring projects about the end and reversal of what once was called the “third wave of democratization.”
The reversal of democratization has had a true global footprint. Not only have democratization processes slowed down or remained stuck, but also hybrid regimes have devolved back into authoritarian directions and autocracies have fortified. The quality of consolidated democracies has degraded as well. EU member states and once consolidated democracies such as Hungary and Poland already appear to have left the democratic development path; the Democracy Index of the Economist Intelligence Unit labels the United States as being a “flawed democracy” since 2016. V-Dem also noted in its 2020 Democracy Report that the United States suffers from substantial autocratization.
However, the democracy ebb, or “the third wave of autocratization,” did not only affect the quality of political regimes. It also had a significant transformative effect on the market of democracy measuring. It amplified already existing methodological criticisms of democracy indices that have dominated the market for a long time, like Freedom House’s Freedom in the World or Polity. Indices that were unable to track the obvious changes in the democracy landscape have lost importance. It also became obvious that there is strong demand both in the academic and policy community for a complex, multifaceted democracy-measuring project that is both devoid of the methodological and conceptual weaknesses of its predecessors and able to capture the complexity of democracy in a way that does not reflect cultural and political bias favoring Western-type liberal democracy. Among the various, complex democracy-measuring projects started between 2005 and 2010, one clearly stands out today and has grown into one of the largest contemporary research projects in political science: Varieties of Democracy (V-Dem).
The book under review, written by Michael Coppedge and coauthors, is an introduction and guide to the Varieties of Democracy project. Because it is not a reflection on V-Dem, but rather a handbook to the project discussing its foundation, conceptual background, and main methodological issues, it is nearly impossible to review the book without effectively reviewing the Varieties of Democracy project itself.
The manuscript was finished in 2018. At that time, V-Dem consisted of 450 indicators, 47 mid-level indices, and 5 high-level democracy indices, totaling up to more than 27 million data observations and covering 201 actually existing or historical polities from the period of 1789–2017. V-Dem 10, released in March 2020, consists of more than 470 indicators, 82 mid-level indices, and 5 high-level indices covering 202 polities from the period of 1789–2019. If one compares these numbers to the 25 democracy indicators of Freedom in the World or to the 7 mid-level and 1 high-level indices of Freedom House’s Nations in Transit, the difference seems astonishing.
Nevertheless, V-Dem’s success cannot be attributed to a single decisive factor. The project’s academic rigor, high degree of complexity, advanced quantitative methodology, impressive logistics, and highly effective fundraising were strictly interlinked and went hand in hand. For example, by 2018 the project had raised more than 26 million USD, and it was working with more than 3,000 country experts.
V-Dem’s uniqueness is also manifested in its data collection and data cleaning methods. Every single indicator is coded by at least five country experts in total isolation from each other. Given that the project is based on 11 surveys organized in four clusters, the number of country experts involved in the creation of a country-year dataset may exceed 20. V-Dem’s measurement model aggregates these ratings into point estimates, also addressing reliability issues like differential item functioning (DIF) or the cross-country comparability of ratings. The statistical and methodological tools addressing the issue of rating reliability and the opportunity to set the confidence level of an expert rating, combined with the unusually high number of country experts involved in the rating exercise, result in outstanding data reliability in comparison with other, less complex democracy-measuring projects.
The desire to avoid a conceptualization of democracy that is either too narrow or too biased toward a Western understanding of liberal democracy led to a conceptual scheme that embraces all seven important academic traditions of democratic theory: electoral, liberal, majoritarian, consensual, participatory, deliberative, and egalitarian. Although present in the conceptual scheme, the majoritarian and consensual approaches are not operationalized and do not have high-level indices, as the other five approaches do. Electoral democracy is considered to be the core concept of democracy; its rating is also incorporated into the ratings of the other four high-level indices. Otherwise, the liberal, participatory, deliberative, and egalitarian components of democracy are considered to be equal. V-Dem waives the aggregation of an overall democracy score in favor of presenting disaggregated scores on the basis of these five high-level democracy indices.
V-Dem’s core strength is definitely its “commitment to disaggregation,” making available hundreds of disaggregated democracy indicators and in that way providing crucial input data for further quantitative democracy research. In this sense, V-Dem is truly a fundamental research project. Regarding conceptualization, the book emphasizes the contested nature of the concept of democracy and formulates a disclaimer stating that “democracy resists authoritative conceptualization,” “others might have made different choices,” and the goal of V-Dem’s “proposed conceptual framework is to provide guidance, not to legislate in an authoritative fashion” (pp. 41–42). Unfortunately, this reserved and reflexive attitude toward conceptualization simply vanishes from the book when it comes to the authors’ assessments of other democracy-measuring projects, especially Polity and Freedom House’s Freedom in the World. These projects are, according to the authors “strictly speaking, not indices of democracy at all” (p. 27) but are instead exercises in “oversimplifying the complex concept of democracy” (p. 25.).
In light of V-Dem’s own disclaimers, the authors could have been perhaps more generous in recognizing others’ freedom to conceptualize as well. This is especially true, because several critical points can be made regarding V-Dem’s conceptual framework as well. For example, although the isolation and identification of the seven democracy principles are done on solid ground, their delineation from each other is often arbitrary. This is especially the case with the electoral and liberal components of democracy, in which certain liberal freedoms, such as freedom of expression and of association, are deemed essential parts of the core electoral concept, in accordance with Robert Dahl’s polyarchy theory (1971), while others are left as parts of the “liberal superstructure” of democracy, which is nice to have but is far from being essential. If V-Dem had conducted a more ambitious conceptualization effort, it could have interpreted Dahl’s polyarchy theory more broadly and put all democracy components under critical scrutiny as to which concept of democracy they really belong.
V-Dem indeed operates with a significantly higher number of democracy components than Dahl did. Therefore, simply copying Dahl’s polyarchy concept to determine the core principles of democracy and relegating all other components to various “superstructure” concepts demonstrates a lack of conceptual ambition. Stating that associated rights, like freedom from torture or political killings, do not belong to the core concept of democracy (p. 33) and are “only” parts of liberal civil liberties represents a missed opportunity to consider the real impact of various indicators on the core concept of electoral democracy. Under the threat of arbitrary torture or political killings, freedom of expression and freedom of association might perish, and the conditions of polyarchy are objectively not met.
It also would have been more ambitious to consider that the liberal component of democracy is still far more essential to the stable functioning of the core electoral concept than are the other principles, but that would have required the introduction of a more hierarchical relationship among the principles and, indeed, could have invited heavy criticism. This initial conceptual convenience is somewhat compensated for by the fact that in its annual democracy reports V-Dem uses the liberal democracy index (LDI) to determine regime characteristics and the extent of autocratization or democratization, which can be seen as a practical recognition of the outstanding importance of the liberal concept of democracy.
Independently from the conceptual debates, during the past several years V-Dem has definitely emerged as the most important provider of quantitative democracy data for scholarly research. For all who contemplate the use of V-Dem data in their work, this book is an essential introduction and guide to the most ambitious, methodologically advanced contemporary democracy-measurement project in the world.
