![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
The Christian Democratic Party (Democrazia Cristiana, DC) constituted the centre of gravity in post-war Italian politics. The party collapsed after 1992, due to long-term changes in Italian society and the eruption of the Clean Hands investigation into political corruption. The DC lost more than half its vote in the 1994 elections: Why did the party lose so much? And where did its votes go? Our answer highlights the effects of modernisation and secularisation as key causes in the demise of the DC vote and in the dispersion of its vote to other parties: parties claiming a lineage from the DC, either truly DC-legacy parties (the PPI, the CCD, the CDU, the UDC and the Margherita), or newcomers on the centre right (Forza Italia and the Northern League), or, finally, the by-products of the party-system’s transformation (the PD and PdL). This article shows that the DC-legacy parties and the Northern League were the most effective in attracting former DC voters. It also demonstrates, once more, the importance of a sub-national analysis by geopolitical regions for a better understanding of Italian electoral behaviour.
Acknowledgements
This article is based on the research made possible by the National Science Foundation grant SBR-94-2281. The Foundation’s support is gratefully acknowledged. The Istituto Cattaneo of Bologna, the Dipartimento di Politica, Istituzioni, Storia of the University of Bologna, and Istituto degli Studi Avanzati of the University of Bologna, the Ministero dell’Interno, and the Istituto Nazionale di Statistica in Rome also provided invaluable assistance.
Notes
1. According to many analysts, with the party system change in 1994 religion lost its centrality as an ideological division while maintaining primacy as a variable influencing voting choice (Bellucci and Heath Citation2011; Cartocci Citation2002; Maraffi Citation2007; Maraffi et al. Citation2011; Sani and Segatti Citation2002; Segatti and Vezzoni Citation2008).
2. Agricultural employment declined from 43% of the economically active population in 1951 to 3% in 1991. Employment in industry and services went in the opposite direction. Italy, one of the most urbanised countries in Europe, saw the proportion of the population living in areas of 20,000 inhabitants or more rise from 54% to 67% between 1951 and 1991. Education, measured by the proportion of the population over 6 years of age with at least an elementary education (five years of schooling from ages 6 to 11) increased from 37% to 98% between 1951 and 1991. There was a consequential decline of illiteracy: in 1951, 13% of the population above the age of 6 was illiterate, but this had declined to 2% by 1991 (see Figure 1).
3. A discussion of the various theories of secularisation goes far beyond the scope of this analysis. In this article, we share the interpretation which sees the process of secularisation evidenced by the declining influence of the Church on societal worldviews and individual behaviour (see Berger Citation1967; Casanova Citation1994; Chavez Citation1994).
4. ISTAT, Statistiche Report, 18 May 2011. This trend has continued. In 2011, 61 percent of the marriages were civil marriages: see Istat, Statistiche Report, 28 November 2012.
5. ISTAT, Il Matrimonio in Italia, nota informativa, 12 February 2007.
6. These data come from the ‘VII Rapporto sulla secolarizzazione’ in Critica Liberale (2011: 18).
7. Not all the Catholic associations underwent decline. On the contrary, some movements – the most relevant example being Comunione e Liberazione (Communion and Liberation) – flourished. The changing membership of the myriad of groups devoted to spiritual concerns, to civil and social action, or to professional activities did not compensate for the massive outflow from the most established organisations in the late 1960s and the 1970s. However, if we include groups and movements not institutionally linked to the Church but staffed by Catholic people, especially in the area of charitable activities, the number of persons involved is quite high, totalling more than 10% of the population (Garelli Citation2007, 108 ff; in this analysis the Scouts, which recruit 200,000 boys and girls, plus the older guides, are not taken into account).
8. Taxpayers have the choice of stipulating that 0.08% of their income tax contributions (otto per mille) are to be paid either to a religious denomination or to the state. Among those who do make the choice (around one-third of tax payers), almost all indicate the Catholic Church. However, since the vast majority (around 60%) do not make any choice, their 0.08% is distributed proportionally among the denominations and the state. This means that the Catholic Church receives almost 90% of the otto per mille contributions.
9. The law, introduced in 1993 and in force until after the 2001 elections, provided for two slightly different electoral systems for the Lower Chamber (Camera) and the Upper Chamber (Senato). Three-quarters (474) of the members of the Camera were elected using the single-member, simple plurality system, the remaining quarter (156) on the basis of proportional representation. Parties presenting lists in the proportional arena needed to overcome an exclusion threshold of 4% in order to be eligible to receive seats. Moreover, the distribution of these seats took place after the application of what was known as the scorporo (literally, ‘unbundling’): in essence, region by region, the votes of a party that elected a deputy in the majoritarian arena were deducted from the votes received by the same party in the proportional arena, the mechanism having been conceived in order to favour those lists which failed to elect deputies in the single-member constituencies. The Senato was elected on the basis of more or less the same mixed system – three-quarters being elected in a majoritarian arena, one quarter proportionally; but the mechanisms governing application of the scorporo were different.
10. The geopolitical division of Italy follows the suggestions of Arculeo and Marradi (1995):
– Nord laico/Secular North: comprises Piedmont, Liguria and Friuli-Venezia Giulia, plus the provinces of Milan, Pavia, Cremona, Varese (in Lombardy) and Venice, Belluno and Rovigo (in Venetia).– Zona bianca/White Area or Northeast: comprises the provinces of Como, Sondrio, Bergamo, Brescia (in Lombardy), Trento (in Trentino), Verona, Vicenza, Treviso, Padova (in Venetia), and Lucca (a Christian Democratic enclave in Tuscany).– Zona rossa/Red Belt: comprises all the provinces of Emilia-Romagna, Tuscany (except Lucca) and Umbria, plus Mantova (in Lombardy), Pesaro and Ancona (in Marche).– Centro/Centre: comprises the all provinces of Latium, Abruzzi, Sardinia, and the provinces of Macerata and Ascoli Piceno (in Marche).– Sud/South: comprises the provinces of all the southern regions (Molise, Campania, Puglia, Calabria, Basilicata and Sicily).The provinces of Aosta and Bolzano are excluded because of the presence of dominant, traditional, regional parties.
11. We also adopted this approach concerning secularisation. While survey data can be of great value, there are well-known problems of validity and reliability concerning religion associated with them. The most common indicator of religiosity – attendance at religious services – seems particularly influenced by over-reporting; moreover, the reliability of over-time analysis is undermined by variations in the wording of questions, with changes in wording eliciting markedly different responses (Rossi and Scappini Citation2010). For example, Pisati’s (2000) overview of survey analyses of Church attendance in Italy since 1956 carefully reports all the discrepancies associated with the various questions. Empirical validation of figures for Church attendance in Venice, by Castegnaro and Della Zuana (2006), demonstrated significant over-reporting. Therefore we adopt civil marriage as a crucial measure of secularisation. Since the 1929 Concordat between Church and State in Italy, municipal officials have recorded all religious marriages in civil records. Alternatively, the marriage can be celebrated in the town hall only, officiated by a member of the local council: this is the civil marriage.
12. Because the measures are across time, and because municipal boundaries changed over time, we construct constant units of analysis. The rule of thumb is that whenever units are merged or divided, the data are aggregated into a single unit. This is why this number is inferior to the standard number of municipalities, which comes to a little more than 8000.
13. The number of second marriages (the Church prohibits religious marriages after divorce), while increasing in recent years, has not significantly affected the increase in civil marriages.
14. The secular parties are excluded from our analysis.
15. Actually, the official merger between the two parties took place a year later, in 2002.
16. Some new entrants were, as followers of Romano Prodi – nominated President of the EU Commission in 1999–part of the Catholic tradition; whereas others, as followers of the former public prosecutor central to the Clean Hands investigations, Antonio Di Pietro, were not.
17. SEM models must be tested for overall goodness-of-fit measured by how well the defined model can recreate the observed data. We employ several tests: first, we test goodness-of-fit for models of modernisation, secularisation and DC vote only. Only the all-Italy model is presented in Figure 2, because of the magnitude of regional variations in the post-DC era. All the models are available upon request. Next, we test the model expanded to include the successor parties in the subsequent Figures. The first set of models – excluding the post-DC parties, for modernisation, secularisation, and DC vote – provide very good fit to the data with the anticipated variations across the areas. The following are the Comparative Fit Index (CFI) and the R2 figures for the DC vote: Italy: CFI, 0.84, DC R2 0.17; Northwest, 0.98, R2 0.22; Northeast, 0.90, R2 0.22; Red Belt, 0.96, R2 0.19; Centre, 0.96, R2 0.19; South, 0.94, R2 0.11. Each area model generates a better fit than the model for Italy, highlighting the distinctiveness and importance of the area variations. For a discussion of goodness-of-fit measures, see Garson (Citation2012).
18. The seminal article by Arculeo and Marradi (Citation1985) provides the milestone for this kind of ‘within area’ analysis. To exclude analysis of Italian politics by area creates the unmeasured relevant variable problem. This condition arises when the researcher omits a key variable that can significantly alter the results. In the study of Italian politics, omitting area from the analysis masks significant differences.
19. More generally, Catholic followers voted disproportionally for the centre-right coalition: 57.4% in 1994, 48.4% in 1996, and 50.5% in 2001 (Sani and Segatti Citation2002, 264–5).
20. To compute the total effects, we recomputed the model treating DM as an exogenous variable to DS. This means that we consider DM as having both a direct effect on the DC vote, as well as an indirect effect on the DC vote via its direct effects on DS and that DS has a direct effect on DC vote. We test the model for collinearity to insure we are not over-determining the total effects coefficient.
Additional information
Notes on contributors
Piero Ignazi
Piero Ignazi is professor of political science at the University of Bologna, Italy. His work focuses on political parties in Europe.
Spencer Wellhofer
Spencer Wellhofer is professor of political science at the University of Denver, USA. His work focuses on transitions to democracy and political parties.