In their recent book Casal Bértoa and Enyedi provide the most comprehensive assessment of the causes and consequences of party system institutionalization in all historical and currently existing European democracies. The volume’s rich theoretical insights and findings will be of interest not only to scholars of party politics but also those interested in democratization, especially given the evidence the book reveals about how party system institutionalization influences democratic survival and the quality of democracy.
Building on Peter Mair’s work, the authors make a solid case for evaluating party systems by focusing on the relation between parties at the governmental arena as the most consequential realm of party competition and cooperation. They argue that the mark of a highly institutionalized party system is the closure of the government formation process: the routinization of interactions between poles of party actors that limits their room for innovation in the composition of cabinets. A closed pattern of competition would be characterized by no or wholesale government alternation as opposed to partial alternation, a familiar formula of party composition of cabinets, and a limited access to cabinet for new parties. In this way, the closure index emphasizes the role of agency of party elites: their loyalty towards previous allies and reluctance to cross ideological divides or other fault lines and their commitment to do so even when voters experiment with new parties.
The theoretical innovation is matched by an equally impressive empirical contribution. The book leverages a unique dataset which allows the authors to track the cabinet alternation patterns, formulas and degree of governmental access via changes of individual ministers across 1347 cabinets from 65 European party systems during 171 years of history. This represents the entire universe of party-based cabinets in European democracies from 1848 to 2019.
The volume is structured in an intuitive manner and makes a great read. The first two chapters introduce the concept of closure and its operationalization, report several validation exercises and present the dataset used in the book. Chapters 3 and 4 include concise accounts of all the 65 currently existing and defunct party systems analysed in the book, sketching common trajectories of change, and discussing the degree to which they fit the party system typology developed by the authors. Chapter 5 compares closure across historical periods and in the three regions of Europe, analyses how the trajectories of closure covary with the party system types and provides a closure ranking of all 65 party systems. The next four chapters explore separately the longitudinal trends, geographical variation and bivariate correlations between closure and four sources of closure, which were also often used by the previous literature as proxies for party system institutionalization: the age and birth-period of democracy, party institutionalization, polarization and fragmentation. Chapter 10 analyses the combined and relative explanatory power of all these factors for closure but also for electoral volatility. Finally, chapter 11 explores the effect of closure on the survival of democracy and democratic quality.
The party system institutionalization index is in itself a major contribution to the empirical study of party politics, providing a methodologically rigorous measure of the stability of the supply side of party politics. This is the long-awaited counterpart for Pedersen’s electoral volatility index which captures the stability of the demand side of politics. I share the authors’ hope that the closure index will be adopted by political scientists as widely as the electoral volatility measure: certainly, the structuring of the dataset makes it very easy to use. Thus, the closure index comes in three variants: a cumulative measure for the entire lifetime of a party system, decade average closure scores and yearly closure scores which are time-weighted to reflect the increased importance of similarity with recent cabinets. This is highly convenient and will enable researchers to appropriately account for party system institutionalization in a variety of historical, time-series or period analyses models.
The volume is a massive improvement on previous attempts to account for the effect of time on party system institutionalization, which neglected defunct cases, did not include Eastern Europe, or focused solely on electoral volatility. The evidence uncovered by the authors strongly indicates the existence of life-cycle effects as opposed to cohort effects. Thus, older party systems tend to be more institutionalized, but this is not due to them being founded in the era of mass parties and strong citizen party attachments – as were many of the party systems which did not survive.
The analysis of countries which have experienced multiple party systems reveals that “as a country moves from one system to another, the tendency is towards more, and not less, closure” (p. 143). This raises a very interesting question which remains unexplored in the book: does party system collapse make political elites wiser or more likely to cooperate in ways inducive to closure or do the new constitutions adopted post-failure include institutional mechanisms that facilitate party system institutionalization? Investigating the former aspect would require a theory of how party elites’ behavioural norms are shaped by the collapse of a system. Regarding the latter, institutional aspect, a provision that is associated with considerably higher government durability (Rubabshi-Shitrit & Hasson 2022) and could change the calculations of party elites regarding wholesale vs. partial cabinet alternation as well as contribute to less innovative cabinet formulas is the constructive vote of no confidence. This very high threshold for cabinet removal is present in party systems with high levels of closure, such as Spain III, Germany II, Hungary, Albania and Belgium. It is true that it is also a feature of currently existing open party systems such as Poland II and Slovenia.
A related, more general question, but of an even higher policy and normative relevance is whether stable party systems can be manufactured by institutional design? While this question is not explicitly discussed, the book’s insights are not overly optimistic: “closure is a relatively stable characteristic of party systems: one observes similar relative positions [in the ranking of closure] irrespective of which period or what length of time is considered” (p. 117), and age of democracy is the most powerful explanatory factor of closure in the multivariate regression (p. 224). Similarly, the null finding regarding the effect of electoral systems on closure does not bode well for proponents of institutional engineering.
Running most analyses at the most aggregate level makes it rather difficult to test several of the theoretical insights developed in the book. The authors convincingly argue for the need to analyse the structure of party competition in terms of blocs and poles, the latter being defined as “parties, or groups of parties, that have the potential to form a government and to present one of the fundamental ideological alternatives of the elections” (p. 14). They propose a new typology of party systems that distinguishes between eight different configurations based on the number of poles and the centrifugal or centripetal tendencies that they induce. Chapter 3 uses this typology to map the currently existing and defunct party systems and illustrates that there is a non-trivial amount of change in the structure of party competition within each party system. Nevertheless, the analytical value of the eight types of configurations is not further probed in the multivariate regression analyses explaining the degree of closure.
In a previous contribution, one of the authors argued that “economic crises… stabilize unstable party systems and destabilize stable ones” (Casal Bértoa & Weber 2019: 234). Unfortunately, this claim remains untested in the current book as the authors claim data on suitable economic indicators was not available for all periods analysed. It would still be highly relevant to analyse even for a reduced sample of cases the effect of economic recession on closure and whether it moderates the relationship between fragmentation, polarization and party institutionalization and closure. This would still be feasible if the unit of analysis would be changed to decades averages or the weighted yearly closure scores.
The survival analysis presented in chapter 11 draws on 282 decade averages to show that, alongside with high levels of GDP, high closure decreases the likelihood of democratic collapse, whereas polarization, fragmentation, party institutionalization, and electoral volatility do not affect it. Would this finding be robust to controlling for variables from conventional models of democratic survival, such as years since transition, democratization wave, exposure to economic recession or war? Similarly, even if such covariates are not available, would the relationship between party system openness and democratic collapse hold if the authors would employ a split population survival model to distinguish between those democracies that survive because they are consolidated and those that are not consolidated but survive because of some favourable circumstances (Svolik 2008)?
One of the most interesting findings of the book is that party system institutionalization is negatively correlated with the quality of democracy in countries with lower levels of economic development. The authors argue that this is due to the weakness in poorer societies of alternative channels of preference representation such as trade unions or social movements that can ensure parties do not become alienated from society and do not succumb to the state capture temptations that closure can enable. Moreover, in wealthier societies, the appeal of the clientelist practices that the non-competitive road to closure can also entail would be, they claim, considerably weaker. Nevertheless, it seems that the effect of the interaction between closure and GDP levels on quality of democracy is evaluated with a cross-sectional regression model in which the unit of analysis is the party system. This reviewer wondered if the result would be robust to rerunning the analysis on a panel set up that would allow to observe whether and how the effect of closure on democratic quality changes as countries become richer.
The qualitative discussion of the Eastern European cases displaying closed party systems and lower quality of democracy and the connection between closure and the ways in which dominant parties are engaged in democratic erosion in the region is highly informative. However, from this discussion of party politics in Georgia, Albania, Hungary, North Macedonia, and Montenegro it also becomes more evident that relying on the proxy of anti-establishment voter support to measure polarisation is rather problematic for Eastern Europe, given the role of radicalized mainstream parties in driving polarization in the region.
The excellent work done by the authors opens several salient avenues of research, beyond those already acknowledged in the conclusion where they rightly point to the likely impact of closure on “electoral participation, corruption, patronage, representation, and accountability” (p. 265). Two can be mentioned. First, the aggregate closure levels and its components can contribute to the research on policy agendas and issue attention stability by helping us understand better the governmental and party systems correlates of policy change. Second, the new data would also allow to investigate whether under-institutionalized party systems are more prone to raise the appeal of and make more likely alternative solutions that further erode party democratic legitimacy, such as technocratic cabinets (Valbruzzi 2020).
Party system closure… is an intellectual tour de force which massively enriches the conventional ways of thinking about party politics. It will most likely become the key academic reference for understanding what drives party system institutionalization and it could also generate considerable debate about how the latter affects democracy. The increased fragmentation, polarization and the loss of electoral dominance of established parties suggest an upcoming upheaval of party systems in Western Europe, even if thus far these systems have maintained relatively high levels of closure and the logic of government formation has continued to be predictable. Judging by the recent German cabinet formation process, the theoretical framework developed in the book seems to also have a fair amount of predictive power, a rare quality for political science work.
