Despite primary elections in Italy continue to be asymmetric – i.e. carried out only by the center-left coalition – their ability to involve the electorate and their growing media impact make it a powerful democratic tool. In this article we study the 2012 Italian primary elections, held by the center-left coalition in order to select the prime ministerial candidate for the 2013 general elections. In particular, we will shed light on three dimensions: turnout, electoral results and competitiveness. We will also take into account the role played by the new candidate selection rule – the two-round system – which will allow us to collect a lot of information about the voting behavior of the selectorate. What has been the turnout level in the 2012 Italian primary elections? Which similarities and differences can be found in the patterns of participation between the first and the second round? Which factors may explain the territorial differences in turnout levels? What have been the territorial patterns of voting behavior for the main candidates? The 2012 primary elections have been more or less competitive with respect to the previous Italian national primaries? We will try to address these questions through the use of a mainly quantitative methodology with aggregate data.

In 2011 Italian local elections we observed high electoral mobility: in Milan, for example, the center-left gained his first-time victory in the Berlusconi era, while in Naples there was a significant split voting in the first round and a huge turnaround between the first and the second ballot. A general research question emerged: are the shifts in the results understandable trough a left-right axis (political nature hypothesis of these elections) or were there cross-cutting mechanisms (local nature hypothesis of the elections with a strong role of personal aspects)? To answer the question we analyze the voting ecological estimates in the three biggest cities involved in 2011 elections: Milan, Naples and Turin. For every matrix we generated the estimates both applying the traditional Goodman model (for the whole city and splitting by district) and the hierarchical multinomial-dirichlet model developed by Rosen, Jiang, King and Taner. The most important result of our study is the strong political polarization of the vote in the two northern cities and a great importance of the local factors in Naples, where only a dominant role of the candidates can make sense of the detected shifts in voting behaviour.

De Sio, L., & Cataldi, M. (2014). Tanto tuonò che piovve: Il risultato delle elezioni. In Terremoto elettorale. Le elezioni politiche del 2013 (pp. 97–128). Bologna: Il Mulino.

GROFMAN, B., CHIARAMONTE, A., D’ALIMONTE, R. D. R., & FELD, S. (2004). Comparing and Contrasting the Uses of Two Graphical Tools for Displaying Patterns of Multyparty Competition. Nagayama Diagrams and Simplex Representations. PARTY POLITICS, 10,...

De Sio, L. (2008). Il secondo motore del cambiamento: i flussi di voto. In Itanes (Ed.), Il ritorno di Berlusconi. Le elezioni politiche 2008 (pp. 57–70). Bologna: Il Mulino.