Going out of the ordinary. The de-institutionalization of the Italian party system in comparative perspective

ABSTRACT

Since 2013, the Italian party system has been in turmoil as the old bipolar structure has been swept away by the emergence of new competitors and skyrocketing voter volatility. Instead of being characterized by stabilization, the 2018 general election continued to show turmoil, with a substantial shift in the balance of power among parties and electoral poles. This article tests the hypothesis that since 2013 Italy has been experiencing a process of party system de-institutionalization, meaning a context in which the interactions of inter-party competition remain unstable and unpredictable over time. To do so, it analyses the patterns of stability and predictability in the three arenas (electoral, parliamentary and governmental) where inter-party competition can be detected, adopting a twofold comparative perspective: a diachronic perspective comparing the 2013–2018 period with the previous Italian republican era; a cross-national perspective (using an original dataset covering 372 elections and legislatures, and 670 governments) comparing Italy since the War with the corresponding period in nineteen other Western European countries. The article paints a rather clear picture: in the period following the 2013 elections the Italian party system is de-institutionalized like very few others in Western Europe’s post-1945 history.

KEYWORDS:

• Party system institutionalization
• volatility
• innovation
• inter-party competition
• Italy
• Western Europe

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. It is a good idea to emphasize the importance of this operationalization, since the use of a thorough set of indicators – and of the corresponding measures which we shall describe during the course of our analysis – is not commonly found in the literature. According to Mainwaring and Scully (1995), party-system institutionalization is the result: 1) of stability of the rules and substance of inter-party competition; 2) of the social rootedness of political parties characterized by a stable ideology; 3) of the legitimacy with which political parties and the electoral process are endowed, and 4) of the importance of party organizations. In fact, even where the concept of party system institutionalization has been carefully broken down into its constituent elements, its empirical analysis goes no further than the stability of party alignments, ideally measured by electoral volatility (Pedersen 1979). Thus, on the one hand, the multi-dimensional nature of the concept is nullified by what is actually a mono-operationalization; and on the other hand, it becomes impossible to distinguish institutionalization from a single change in the party system, which in fact is often empirically analysed by recourse to the index of electoral volatility. As we shall see later, we ourselves use this very same indicator, albeit not this one only.

2. This approach, with an application to the Italian case, has also been adopted by Calossi and Cicchi (2018).

3. Electoral volatility (EV) is calculated as follows: EV = $\frac{{\sum }_{i=1}^n|v_{it-}{v}_{i\left(t+1\right)|}}{2}$, where n is the number of parties, and v_i represents the percentage of votes obtained by that party at time t and at time t + 1. It should be pointed out that EV measures the change in the party distribution of the vote and not individual changes in votes. At the very most, it could be said that with the same body of voters, the value of EV expresses the minimum degree of individual change present in the system. Thus an EV of 10 implies that at least 10% of voters have switched votes between two consecutive elections. In addition to this, volatility also takes account, to some degree, of changes in the electoral supply (the parties standing for election), since it reflects the shifts in the votes of those who, at a given election, voted for a party no longer competing in the subsequent election.

4. To 2019, the most volatile elections in Western Europe were ranked as follows (the value of EV is shown in brackets): 1) Greece May 2012 (48.5); 2) Spain 1982 (43.8); 3) France 2017 (40.7); 4) Italy 1994 (39.3); 5) Italy 2013 (36.7); 6) Spain 2015 (35.5); 7) Iceland 2013 (34.7); 8) Netherlands 2002 (31.3); 9) Iceland 2016 (31.1); 10) Ireland 2011 (29.6); 11) France 2001 (27.8) and 12) Italy 2018 (26.7). It is interesting to note that ten of the twelve elections in question were held during the last twenty years, confirming the overall increase in the levels of volatility present in Western Europe. The above figures were taken from Emanuele (2015b).

5. The definition of a ‘new’ party is a hotly debated issue among party-politics scholars. The operational definition used here is the one suggested by Bartolini and Mair ([1990] 2007): mergers and splits are not conditions for the foundation of new parties, and the latter include only those that are organizationally new. See Emanuele and Chiaramonte (2018) for a review of the scholarly debate on this topic.

6. The analysis of electoral innovation starts from the third democratic election (in our case, the 1953 election), that is, after the initial institutionalization of the party system founded following the end of WWII. On this point, see Morlino (1998).

7. PV = $\frac{{\sum }_{i=1}^n|s_{it-}{s}_{i\left(t+1\right)|}}{2}$, where n is the number of parties, and s_i represents the percentage of seats obtained by that party at time t and at time t + 1.

8. As with electoral innovation, those parties originating from mergers or splits are not considered to be new. Once again, the index is calculated starting from the third democratic election, held in 1953.

9. The innovation score of 0.3 was due exclusively to the two parliamentary seats obtained by +Europa. In calculating innovation, as with volatility, no account was taken of the results in the constituencies for citizens resident abroad.

10. This is how Mair (2006) calculates the index of governmental alternation. For a different calculation of ‘governmental turnover’ based on parliamentary seats rather than on governmental seats, see Ieraci (2012).

11. GV = (50- |$\frac{{\sum }_{i=1}^n|m_{it-}{m}_{i\left(t+1\right)|}}{2}$ – 50|)*2 where n is the number of parties, and m_i represents the percentage of ministerial posts held by that party at times t and t + 1.

12. As the graph shows, there has never been a perfect political alternation, given the considerable number of ministerial posts held by technocrats (27.5% of all ministerial posts between 1994 and 2013). The latter are considered to represent continuity with the past for the purposes of calculating volatility. This decision is designed to emphasize the differences between those countries where there is total alternation of diverse political parties (e.g. the UK or Malta, or less commonly, Ireland and Sweden) and where a model of party government (Blondel and Cotta 2000) operates, and those, such as Italy, where alternation is often diluted by a recourse to technocracy.

13. These comprised Italia dei Valori (Italy of Values), Rifondazione Comunista (Communist Refoundation), and the Rosa nel pugno (Rose in the Fist) alliance, the latter being in turn part of Romano Prodi’s centre-left governing coalition, L’Unione (the Union) during the 2006–2008 legislature.

14. Beyond the case of the M5s, among the parties with a prominent electoral breakthrough, only La Republique En Marche in France gained immediate access to government.

15. Each box plot is to be interpreted as follows: the box includes 50% of values (the first quartile is represented by the bottom side of the box, the third quartile by the top side, and the line within the box identifies the median). The two whiskers at the top and the bottom of the box include all remaining values that are not outliers, while the latter (defined as those observations that are at least 1.5 times lower than the first quartile or 1.5 times higher than the third quartile) are represented by the dots.

Additional information

Notes on contributors

Vincenzo Emanuele

Vincenzo Emanuele is Assistant Professor in Political Science at Luiss, Rome. He is Co-Chair of the Research Network on Political Parties, Party Systems and Elections of the Council of Europe. His research, focusing mainly on elections and party system change, has appeared – among others – in Comparative Political Studies, West European Politics, Party Politics, Government and Opposition, and South European Society and Politics. His monograph, Cleavages, institutions, and competition. Understanding vote nationalization in Western Europe (1965–2015) has been published by Rowman&Littlefield/ECPR Press.

Alessandro Chiaramonte

Alessandro Chiaramonte is Professor of Political Science at the University of Florence. He is a founding member of CISE (Italian Centre for Electoral Studies). He is author or editor of several books on elections, electoral systems and Italian politics. He has published articles on West European Politics, Party Politics, South European Society and Politics, Government and Opposition as well as on the main Italian Political Science journals.