Mayors and votes in Rome in the “Second Republic”: an in-depth analysis

Introduction: The year 1993 marks the transition from the First to the Second Italian Republic—a moment of profound political, economic, and social transformation that clearly divides the 1990s into a “before” and an “after.” At the centre of this shift is Law No. 81, which introduced the direct election of mayors and became a cornerstone of a paradigmatic change in local governance. Thirty years later, this article seeks to take stock of the transformations that have taken place over this period, offering a longitudinal analysis of electoral behaviour and political participation in the city of Rome.

Methods: The study draws on data provided by the Ministry of the Interior and the City of Rome. These data concern voting behaviour and electoral turnout from the 1990s to the present. They also cover the political spectrum (coalitions, lists, and parties), the role of the mayor, and the composition of the city council, with particular attention to the presence of women and the socio-political characteristics of the municipal—and by extension, administrative—classes.

Results: The study focuses on the nature of electoral competition and the evolution of political power across successive electoral cycles. It pays particular attention to the succession of mayors, the dynamics of their popular support, and how this support has been redefined, reshaped, and reduced over time. The analysis also examines changes in the political class, the composition of the city council and executive bodies, and how today’s “Roman politician” differs from that of the 1990s, including in terms of gender representation. Finally, the article reflects on the structure and legacy of Law No. 81, acknowledging its role in reshaping local government, but also questioning the current strength of its founding pact—namely, the one between citizens and the mayor.

Discussion: The study reveals a marked decline in voter turnout and a progressive erosion of popular support for the mayor. This trend points back to Law No. 81, which—although initially heralded as revolutionary for local governance—now appears somewhat outdated, or at least in need of revision. The founding “citizen–mayor pact” is increasingly fragile, and may be approaching an irreversible decline.

1 Introduction

If we had to pinpoint the year in which Italy transitioned from the “First” to the “Second” Republic, the choice would surely fall on 1993, when sweeping changes definitively marked out a before and an after. The Mani Pulite “Clean Hands” investigation that would overwhelm the political system began in 1992 and ended with the “Cusani trial” in 1994 (Giglioli et al., 1997). The input provided by the investigation turned 1993 into a year of power-plays and drama and opened up the way to a different political arrangement. The political parties involved left the field in various states of distress; Christian Democracy (DC), which had led the country for almost 50 years, stood for the last time in the elections of 1992, and thereafter broke into fragments as it tried in vain to regain its lost central position (Lombardo, 1984). The Socialist Party (PSI), which for the previous decade had been a symbol of a “free-wheeling” idea of governance, and a new style of leadership in the figure of Bettino Craxi, also suffered a breakup leading to its present-day virtual extinction. The Italian Social Movement (MSI) emerged “clean” from the trials and at its 1995 Fiuggi convention, was reborn as first, the National Alliance (AN) and then later as Brothers of Italy (FdI). In 1993, however, it was already leading one of the two groups, or poli, facing each other in the Rome administrative elections; and in 1994 it became a pilaster of the centre-right alongside Berlusconi’s Forza Italia (FI). The Italian Communist Party (PCI) fell apart with the fall of the Berlin Wall and at a pivotal congress in 1991, transformed itself into the Democratic Party of the Left (PDS). It immediately became the leader of the progressive wing, and in 2007 was the “major shareholder” in the Democratic Party (PD). The 1991 breakup also led to the formation of traditionalist groups, led by the Communist Refoundation Party (PRC). On the right, the Northern League, created in the previous decade under the leadership of Umberto Bossi, survived the investigations with only a few scratches, and entered the new Second Republic as the longest lasting political group in parliament, despite being the “youngest” only a few years previously.

After 2 years of upheaval, by 1994, the political system already looked radically different, with changes made at organisational, procedural and also ideological levels. Even the relationships between the centre and the periphery were redefined, towards a “politics without territory” (Diamanti, 2003a, 2008). The old ruling class followed the fate of the parties, in a downwards trend that opened their ranks to people with no party history. After the Clean Hands campaign had ground to a halt, in the Second Republic civil society became the mainstay of legitimacy for a system that was seeking to redefine itself. From this system emerged Silvio Berlusconi, the founder and father-figure of Forza Italia, who became the main protagonist of the ruling class of the last 30 years (Ignazi, 2014), and from this same background actors appeared who were preparing themselves to govern cities and regions.

Everything dates back to 1993, and there we can begin our analysis of the situation in Rome, whose administration was fully involved in the dynamics of change. This article deals with politics at the local level, which exists in symbiosis with the political centre (Cedroni and Fava, 2013, 2022). Examining politics at this “lower” level (della Porta, 1999) allows us to analyse different aspects of institutions simultaneously, including the reforms that redefine the rules of the game (March and Olsen, 1992) and affect the political processes, within a context that has intertwining economic, social, cultural and political dimensions.

1993, however, was the symbolic year in a decade of reforms, including the electoral laws for the Chamber and Senate, the new mechanisms governing the election of regional presidents, the Bassanini laws which, between 1997 and 1999, would regulate the process of administrative decentralisation. Until the controversial modification of Title V of the Constitution in 2001, the reform “par excellence” remains, however, that introduced under Law 81, in March 1993, which ushered in a reorganisation that created a new electoral system in city elections, restyling the form of municipal government (Vandelli, 1997; Agosta, 1999). This reform was hailed as “a success” (Piselli et al., 2012), and “the most accomplished of the Italian transition” (Fabbrini, 2001). Not only 1993, but also Law 81, therefore, had a clear and distinct “before” and “after.”

Previously, city administrations had an average duration of 2 years (Cazzola, 1991), elections were based on a proportional system and the political framework was controlled by the parties (Tarrow, 1979; Canzano, 2007, 2012) who chose the candidates, defined alliances and made nominations (Di Virgilio, 2006). The office of mayor represented a “career stage” that had the party as its centre of gravity and the mayor was first among equals (elected by equals) (Bettin Lattes and Magnier, 1997). He was a mediator at the helm of councils that were not very diverse—whose members were at the head of “fiefdoms” bent on guaranteeing friendly factions—characterised by incremental decisionmaking styles (Fava, 2002) and by low political responsibility. In 1990, Law 142 had introduced innovations that, as well as guaranteeing statutory autonomy to municipalities, sought to reduce the centrality of the council, redefining its power relationship with the executive (Segatori, 2003), but did not interfere with the electoral mechanisms and the substance of the form of local government itself (Di Virgilio, 2006).

All this changed with the Mayoral Reform introduced in Law 81, which aimed to strengthen the figure of the mayor elected by the citizens, recognising a “direct” legitimacy that elevated him to a first above unequals. In municipalities with more than 15,000 inhabitants the mayor was elected by an absolute majority, in the first or second round after a run-off between the two main contenders. The council was created on a proportional basis and the lists supporting the winning candidate could have (with some exceptions) 60% of the seats¹ (Camarda, 2018). The local government was already defined at the closing of the polls and enjoyed predefined stability. The municipal council was weakened and less politically attractive, filled initially at least with actors of secondary importance.

The executive council was stronger and more autonomous, and the mayor, who had the power of appointing (and dismissing) the councillors, was transformed into a monocratic figure who was set at the summit of the executive and was the leader of the majority that supported him (Vaciago, 1999). A new democratic accountability has thus been created, in which responsibility rests directly with the mayor (Piselli et al., 2012). This led to a style of governance based on different dynamics, methods and decision-making styles, networking with civil society, interest groups, and the citizenry. People’s perception of the effectiveness of the administration improved, political obstacles decreased, but bureaucratic impediments increased (Catanzaro et al., 2002).

Popular investiture created the emergence of actors who were strong, recognised, customised and politically positioned; the “first leader” with a group of trusted followers established in key positions who became the “mayor’s party” (Gaspari et al., 2009).

In the 1993 elections, figures emerged who presented an image that was intentionally different. Some came from civil society, such as Castellani in Turin, Illy in Trieste, Sansa in Genova. Others with stronger political backgrounds, but who, in their greater autonomy from the parties, were perceived as a positive change, included Cacciari in Venice, Orlando in Palermo and Bianco in Catania, as well as, obviously, Rutelli in Rome. There was however no lack of individuals whose careers had been made entirely within parties, such as Vitali in Bologna, De Luca in Salerno and Bassolino in Naples.

Given their new visibility and their redefined role, people began talking about the “era of mayors” (Diamanti, 2003b), or the “great political novelty of the 90s” (Mancini, 1997) involving mayors who were mostly from the centre-left; but by the end of the decade and in the years following, centreright figures were increasingly elected, including Guazzaloca in Bologna, Tosi in Verona, Gentilini in Treviso, Moratti in Milan and Alemanno in Rome.

Quite soon, however, the “era of mayors” ran out of steam, as coalition networks grew and old-style councillors reappeared; the city councils became once more places of mediation and political compensation (Vandelli, 1997; Burroni et al., 2009). The connection to the parties became once again the key to success, from Chiamparino in Turin to Russo Iervolino in Naples or Poli Bortone in Lecce, and, certainly, Veltroni first and Alemanno later in Rome, as well as Marino, Raggi and, lastly, Gualtieri.

In our analysis of 30 years of Roman politics, we need to look closely at the different stages, at the innovations, the transitions and above all, the consensus crisis that led to the much reduced role of the mayor today. We need to examine how the type of political competition also changed; more or less bipolar until 2013, then double-sided in 2016, them disbanded in 2021. The focus was now on the vote, on the voters, on the competition, on the actors and the outcomes.

2 Rome 1993–2021: 30 years of voting and participation

From the first municipal elections under Law 81 to today, those entitled to vote (EV) have always been around 2, 300,000.² The actual data however tell a different storey. If the eligible voters can be plotted on a horizontal line, the same is not true for the actual voters (V) who went from 1, 825,000 in 1993 to 1,145,000 in 2021.³ Vote dispersion involved 679,000 voters (−37.2) and the turnout (TO) dropped from 78.7 to below the majority threshold at 48.5. This means that in 30 years more than one voter in three joined the ranks of the abstainers. It should be added that in 1993, in the run-off, the number of voters remained more or less the same as in the first round, with an increase of only 26,000 (+1.8), while in the following rounds, participation dropped vertically. And if in 2001 and 2008 the decline can be attributed, in part, to the concurrence between administrative and political elections (which “increased” the turnout in the first round), in the other rounds the decrease was unambiguous. Between 1993 and 2021, there was a loss of 890,000 voters (−48.1). In 2021, 185,000 voters disappeared (−16.2), the turnout collapsed to 40.7 and voters fell below 1 million for the first time.

Every electoral system has psychological and cognitive effects on the voters and their relationship with the vote (Sartori, 1995; Baldini et al., 2000; Baldini and Pappalardo, 2004).

In the first elections under Law 81, votes in favour of the mayoral candidates were 1,731,000, which gradually decreased to 1,110,000 in 2021. A significant loss equal to 621,000 votes, more than a third (−35.9).⁴ In the last municipal elections of the First Republic held in 1989, the list vote (LV), i.e., for the parties, involved 95% of the voters, which was much the same percentage of votes for the mayor (MV) registered in 1993. As a result, the “cross on the symbol” declines, involving only 70% of the voters, with a decrease of 499,000 (−27.9). The LV lost a further 273,000 (−21.2) in the following 30 years,⁵ but its overall importance to voters can be seen to be strengthened, slightly at first, then after 2008 it is more sustainable, approaching 90% of voters in 2016 and 2021. If we cross-check the LV and MV data, in 1993 the former is 75% of the latter; in 2021 this becomes more than 90%. Voters are keeping their vote for the mayor, but at the same time are paying more attention to the party vote (Table 1).

Table 1

Table 1. Votes and voters 1989–2021.

3 The offer and political choice

Law 81 introduced something new into the political system and opened up a deep rift between the past and what followed. The last three decades, however, have presented a picture not of stability but of a system in continual flux, tending towards a form of local, minority democracy.

3.1 The coalitions

In 1989, the Romans went to the polls with a rather simple electoral system, both in the methods of voting and its effects. Law 81 made it more complex and the voters had to acquire a new mindset (Carrai, 1997). They could choose the mayor and express a preference for a party (even one different from that of the mayor). We went from a vote with one dimension (party) to a three-dimensional one (mayor-coalition-party). The choice of the mayor represented the “first vote”. The coalition “at the ballot box” was an invention of Law 81 and was created when the previous system, based on the “post-electoral” coalition, was abolished.

In 1993, in the midst of the upheaval, the alliances struggled to define themselves. Four blocks entered the competition; The first was the progressive camp (37.2), which included four lists led by the PDS (18.2), the Greens and Radicals, which chose Francesco Rutelli as candidate for mayor (39.6), who was then elected in the run-off (53.1). A second block of the right (33.4) included 2 lists, centred on the MSI (31.0) and on the figure of Gianfranco Fini (35.8), its national leader. A third of the centre (14.2), with 4 lists, nominated Carmelo Caruso (11.4) and was supported by Christian Democrat forces desperately seeking their own focal point. Finally, a fourth of the left (8.0), (3 lists), with RC nominating Renato Nicolini (8.3) as a candidate. The competition ended up limited to the first two coalitions,⁶ the winning progressives and the right-wingers, who obtained 70.6% of the VL. The two “first” parties together, however, did not reach 50% of the VL and with the losing MSI managing to almost double the votes of the winning PDS.

In 1997 the fight was between two coalitions only; the centre-left (57.3), (10), combining its three components (centre, centre-left and left) seeking (and obtaining) the reconfirmation of Rutelli (60.4), and the centre-right (38.6), (5), which grouped together An, Fi and part of the former DC, nominating Pierluigi Borghini as mayor. This time the main coalitions increased in size but the lists and candidates decreased (8, half compared to 1993). Bipolarism was strong, but not competitive, and the two factions shared over 95% of the votes. The two-party opposition was also decidedly weak, with An (24.1) and PDS (22.0) barely exceeding 45% of the votes. Once again the “leading” party was part of the losing coalition.

In 2001 the administrative elections were held at the same time as the political elections. The number of lists and candidates increased, and the competitive coalitions were still the centre-left (45.5), (7) and the centre-right (49.6), (11). The former nominated Walter Veltroni (48.3), who then won (52.0). The latter focused on Antonio Tajani (45.5), representing their second party, FI (19.2), while the first was An (21.0). Bipolarism was strong and competitive, and the two coalitions obtained 95% of the votes, with the loser winning in the first round, just a few tenths shy of 50%. The two-party confrontation was, however, rather weak, with the two major forces (An and FI) both from the centre-right and the DS (the winner) in third place (17.6). In 2006 the outgoing mayor was confirmed (61.4), beating Gianni Alemanno (37.1). The competition was still bipolar, but not competitive. The centre-left (61.4) and the centre-right (37.0) were crowded with lists (12 and 14). The bipartisanship remained weak, with the first party of the centre-left, the Ulivo (33.8), finishing above the total of An (19.5) and FI (10.2), the main members of the centre-right.

After Veltroni, the next secretary of the Democratic Party, resigned, in 2008 the Romans returned to the polling stations; this time, the municipal elections were held alongside the national political elections. The turnout was lower than 2 years earlier, the competition was still bipolar, and the two main alliances obtain 87.6% of the votes. In the first round, Rutelli (45.8) and the centre-left (47.9) were victors; the left was led for the first time by the Democratic Party (DS), recently formed in the “loveless marriage” between the DS and the Margherita party (Fava, 2013). The centre-right stuck at 39.6 and its candidate at 40.7. Everything changed however in the run-off with Alemanno (53.7) managing to beat Rutelli (46.3), down 82,000 votes compared to the first round.⁷ For the first time the two-party competition was real, PdL (36.6) and PD (34.4) neck and neck and capable of taking more than three quarters of the LV.

In 2013, the centre-right (31.7), (6 lists) re-nominated the outgoing Alemanno (30.3), while the centre-left (42.6), (6), chose Ignazio Marino (42.6), who won in the second round (63.9). On the margins were a single-party coalition led by the Five-Star Movement (M5s) (12.8), with Marcello De Vito (12.4), and a centrist group (7.8), (2), led by Alfio Marchini (9.5). The bipolarism was weakened and the two main groups together did not gain beyond three quarters of the votes (74.1). Here, the novelty was the M5s which opened the way to the rise of the “two camps” situation, destined to come about 3 years later. The PD was still first (26.3) ahead of the PdL (19.2). Together, however, they did not reach 50% of the votes and the two-party competition returned to pre-2008 levels.

The ailing Marino administration came to an end prematurely and in 2016 the polls reopened. The offer on hand became more complicated, with four coalitions above 10%. The first, “single-party” group (35.3) was led by the M5s and Virginia Raggi (35.3), who then became mayor (67.1). It was followed by the centre-left (25.4), (7), which nominated Roberto Giacchetti (24.9), then a rightwing coalition (19.6), (5), led by FdI and Giorgia Meloni (20.6), and a centre-right coalition (11.3), (8), aligned around Marchini (11.0). The competition redefined itself and became two-dimensional (see Schadee et al., 2019), with one group articulating the left–right axis and the other emphasising the breach between the representatives and those they represent. This transformation, national and local, was taking place in a growing anti-political, populist climate, characterised by a crisis in representation, a free-flowing vote, and voters becoming an amorphous mass (Revelli, 2017, 2019), who were deserting the logic of a left–right distinction and were moving towards a “neither left nor right” position (see Inglehart and Klingermsnn, 1976). Even in the 2016 round the two-party competition was weak, with the party that epitomised the new inclination, the M5s (35.2), well above the sum of the votes of the PD (17.2) and FdI (12.3), the two main parties in the traditional political axis.

The last round took place in 2021; the left–right dimension became the dominant factor once again (albeit weakly). There were four coalitions above 15%; the centre-left (27.9), (7 lists), with the PD (16.4) as the linchpin with its candidate Roberto Gualtieri (27.3), who then became the winner (60.1); the centre-right (31.4), (6), led by FdI (17.4), with Enrico Michetti their standard-bearer (30.1); a centre coalition (19.1), with a single component, its candidate Carlo Calenda (19.8); and, finally, a populist grouping (17.7) led by the M5s (plus 5 independents) with the outgoing mayor (19.1). Although the competition was ranged along traditional lines, it was the least bipolar so far, due to the high amount of fragmentation,⁸ where the two main groupings (Calenda and PD) collected just over a third of the votes. For the first time, the “civic list” of independents proved to be a major force (Table 2).

Table 2

Table 2. Vote for coalitions and parties (1989–2021).

3.2 The mayors

Between 1993 and 2021 there were eight rounds of elections, and the election of six different mayors, including only one woman (Raggi). There were 121 candidacies for 106 candidates, and only a small number entered the municipal assembly (14.9).⁹ There were 26 female candidates; 21.5 among the candidacies and 9.1 among those elected to the council (2 out of 22).

Despite the continuing decline in voters, candidacies were on the rise, from 13 per election between 2001 and 2008 to 18 in the later period (2013–2021), with the maximum reached in 2021 (22), but with an offer that was increasingly less composite, weaker and fragmented.

In this 30-year period, preferences for the elected candidates were approximately 95% of the mayoral vote (MV), showing a slight increase in the later rounds. The growth in candidacies did not seem, therefore, to affect the weight of the vote for those elected. Something changes, however, when attention is focused on the candidates in the runoff.

In 1993, the two “top candidates” obtained three quarters of the MV, reaching over 90% in the three subsequent rounds, and up to 98.5 in 2006. After 2013, the picture changed and the strength of the two candidates weakened, falling to 60.2 in 2016 and 57.2 in 2021, when the sum of their consensus (635,000) did not reach half of that of their challengers in the first elections under Law 81 (1,304,000).

If we focus our attention on the mayors themselves, we find that in 1993, Rutelli won in the second round after gaining almost 700,000 votes in the first, then increasing by 300,000 in 1997 when, at first count, he almost reached a million. The mayor effect eased off in 2001, but Veltroni’s total was, however, 800,000, rising to 927,000 in 2006. From here there was a substantial drop, to 675,000 votes in 2008, 513,000 in 2013, 416,000 in 2016, and 300,000 in 2021. The sum total of the votes of the winners in the last four contests (2008–2021), failed to reach 60% of those gained in the first four (1993–2006): a veritable collapse from 3,396,000 to 1,949,000.

Gualtieri won in 2021, obtaining 300,000 votes in the first round, but which was less than half of what Rutelli gained in 1993 (685,000), 30% of what he obtained in 1997 (984,000) and less than 40% of the votes of Veltroni in 2001. In his weak showing, Gualtieri won with a package of votes that was equal to the increase obtained by Rutelli in the transition between “93” and “97”. He also won after coming in second in the first round.¹⁰ He was not only the mayor least rewarded by the citizens, but also the one with the fewest votes compared to all the second-placed mayors in the eight rounds.¹¹ (Table 3).

Table 3

Table 3. Candidates, mayors, votes (1993–2021).

Comparing personal MVs with those of the supporting lists, Rutelli surpassed his own coalition by 205,000 votes in 1993 (+42.9) and by 248,000 in 1997 (+33.8),¹² while Veltroni was ahead by 186,000 in 2001 (+30.0). From here the two curves (MV and LV) continuously realign. Differences drop to 125,000 in 2006 (+15.6), to 69,000 in 2008 (+11.4), to 79,000 in 2013 (+18.2), to 41,000 in 2016 (+9.7), and to 17,000 in 2021 (+5.9).

3.3 Parties and lists

The LV in 1993 was 74.4% of the MV, while in 2021 it rose to 91.4. In the first round after the introduction of Law 81, it was 70.6% of the total voters (E), and became 88.6% in the last round. It was also 55.6% of those entitled to vote (EV) in 1993 and 43.0% in 2021,¹³ so there was an upswing in voting for parties, if only in an increasingly narrow field. In 1989, 23 lists were entered, followed by 25 in 1993, 32 in 2001, 40 in 2013 and 38 in 2021; in this crowd were dozens of civic listings,¹⁴ which were often irrelevant from the point of view of consensus.¹⁵ In 30 years, there were 255 lists, of which only 83 obtained seats (one in three), obtaining approximately 90% of the available votes.¹⁶

Out of all of this, a distinction must be made between parties (106)¹⁷ and independent civic lists (149), four parties for every six civic lists. Leaving aside the first two elections, with their tendency towards swings in both directions, between 2001 and 2021 the civic lists went from 13 to 25, doubling the votes and their electoral weight rose from 16 to 40%. There was also another distinction to made between the civic lists tout court, and the “personally” and the “politically” oriented candidates. Between 2001–2008 and 2013–2021 the former grew from 10.3 to 15, the second from 3.3 to 4 and the third from 3 to 6.3, while the parties dropped from 15.7 to 12. Overall the Civic Listings went from 585,000 to 975,000 votes, while the first type increased from 91,000 to 233,000 (+155%, from 2.2 to 7.2 of the LV), the second from 457,000 to 587,000 (+28%, from 10.9 to 18.2) and the third from 36,000 to 155,000 (+330%, from 0.9 to 4.8), while the parties plummeted from 3, 600,000 thousand to 2,250,000 votes (−38%, from 86 to 69.7) (Table 4).

Table 4

Table 4. Votes and types of lists (2001–2021).

In the 2001–21 time frame, lists (parties included) numbered 209. Distributed along the left–right axis, the most crowded area is the centre-right (47), followed by the right (43), centre (38), centreleft (31), left (18) and populist camp (8). After 2008, the growth was mainly on the right and in the populist camp, but also in the centre-right and on the left. Decreases took place in the centre-left and in the centre where, however, the civic lists increased. After 2001, list votes were around 7,400, 000. The centre-left civic lists obtained 2.5 million votes (33.6) exceeding their centre-right peers (19.1) by over a million preferences. They were followed by those on the right (17.1) who reached almost1,300,000 votes, the centre (914,000, 12.3), the populist camp (770,000 10.4) and the left (524,000 7.1).

Between 2001–2008 and 2013–2021 there was a drop of almost 1 million LV (−23%) and the centreleft was the main receptacle, despite losing approximately half (−48.7). The centre-right lost two thirds (−64.4), while, in line with the decrease in voters, the centre (−26.3) and left wing lists (23.8) had around a quarter fewer votes. The populist lists, a novelty of the previous decade, obtained almost 750,000 votes (23.2), becoming the second political force after the centre-left, who managed to stay ahead of the right-wing lists, which, despite abstentionism, exceeded 640,000 votes, remaining in line with the votes of the first period (+2.1) (Table 5).

Table 5

Table 5. Votes and political areas (2001–2021).

In the 30 years after Law 81, five parties were able to exceed 30% of the votes: the PDL and the PD in 2008, the Ulivo in 2006, the M5S in 2016 and the MSI in 1993. Only the first two managed to go beyond the threshold of 500,000 preferences, both of them in in 2008. In 1993, the MSI was the leading party, and became the sole collector of votes from the right and centre-right. It obtained 400,000 votes (31.0), which fell to 309,000 in 1997 (24.1), when AN took over, and then to 254,000 in 2006 (19.5) and to just 60,000 in 2013, the year in which Brothers of Italy (Fratelli d’Italia, FdI) entered the scene. Compared to the MSI, the votes lost in 2013 were 339,000 (from 31.0 to 5.9).

From their lowest point, the climb began, leading to 146,000 in 2016 (12.3) and to 177,000 in 2021 (17.4). FdI, however, were far from the 309,000 preferences obtained by An in 1997 (24.1).¹⁸ In the same year Forza Italia (on the CDU list) participated in its first Roman administrative elections and obtained 129,000 votes (10.9), which went to 260 thousand in 2001 (19.2), then returned to the levels of “97” (133,000) in 2006 (10.2). In 2008 it was absent, as it formed part of the People of Freedom (PDL), which immediately skyrocketed to 560,000 votes (36.6), losing, however, 364,000 in 2013 (19.2). Once the PDL experience was over, FI reappeared in 2016 obtaining 51,000 votes (4.3), which then fell to 36,000, its lowest point, in 2021 (3.6).¹⁹

The PdS/DS, the more moderate heir of the PCI, obtained 234,000 votes in 1993 (18.2), which became 282,000 four years later (22.0) and then fell to 239,000 in 2001 (17.6). In 2006 the Ulivo brought together DS and Margherita and reached 442,000 preferences (33.8). Two years later the PD entered the field, gaining its maximum of 521,000 votes (34.0)²⁰ in its first outing, then suffering decline to 268,000 votes in 2013 (26.3), 205,000 in 2016 (17.2), 166,000 in the last round (16.4). In 2021, it lost over 350,000 votes compared to 2008, or well over double its last effort.²¹

In 1993, the DC in its last appearance obtained 155,000 votes (12.0). Since then, in 2001 the Christian Democrat diaspora, including Margherita, UDC and European Democracy obtained 185,000 votes (13.7), but dropped to 124,000 in 2006 (9.5), to 33,000 in 2016 (2.8) and to 10,000 in 2021 (1.0). The socialist-inspired lists had 30,000 votes in 1997 (2.4) and 2006 (2.5), but moved towards extinction in 2008, obtaining less than 3,000 preferences in 2021 (0.3).²² This was also the fate of the Radical Party, who, with their Pannella list received 45,000 votes in 1993 (3.5), then gradually slipped, under various names, towards disappearance in 2021.

After the Fiuggi Convention, the neo-fascist right always managed to keep its tricolour flame alight, going from 29,000 votes in 1997 (2.3), to 52,000 in 2008 (3.4)²³ then to 24,000 in 2016. Over the years it has included within its ranks Msi-Fiamma tricolore, Fronte nazionale, Alternativa sociale, Forza Nuova, and Casapound, who were all absent in 2021. Neo-fascism however remained a latent force and reached a minimum with only 1,500 votes (0.15).

The left-wing lists won 823,000 votes, over a quarter coming from civic formations, which however contributed as much as 51% in the last three rounds. The vote was split between communist lists and others linked more to environmentalism and other activist ideas, with the latter capable of providing around 45% of the total votes and three-quarters (76.3) of those in the last period. The weight of the civic lists and those of non-communist origin was therefore growing. From the first to the second decade of 2000, 70,000 votes were lost (−23.6). Also, between decline and continual splits, the narrative of the left can be understood by simply observing the “first parties” of the area in the various competitions: 112,000 votes to RC in 1997 (8.8), 64,000 for Sinistra Ecologia Libertà in 2013 (6.3), and 20,000 to Sinistra civica ecologista in 2021 (2.0).

The entry of the Five Star Movement (M5S) on the scene muddles the picture by moving the contest on to the represented-representatives axis. First in 2013, when it gained 131,000 votes (12.8), which rose to 420,000 4 years later (35.3) then fell to 112,000 in 2021 (11.0) (Table 6).

Table 6

Table 6. Votes and parties (1993–2021).

From 1993 to 2021, the Civic Lists won 2 million votes (20.2 VL), 1.2 million came from personal candidates (59.5), and grew by 130,000 preferences (+28%) in 2000, despite the LVs being almost a million less (−23%).

The first personal list (at least one bearing the name of a politician, the “Pannella” list) was already in existence in 1993. Four years later we find the first with an explicit reference to a candidate, “Rome for Rutelli” which obtained 90,000 votes (7.0). In 2001 there were three; the Veltroni list obtained more than 146,000 votes (10.9). A list linked to Antonio Tajani settled at 37,000 (2.8), and the Bonino list got 14,000 votes (1.1). In 2006 there were two, both linked to Veltroni and, together, obtained 138,000 votes (10.5). Two years later there were five, two in support of Rutelli with 82,000 votes in total (5.4), one for Alemanno with 10,000 votes (0.7) and two not linked to mayoral candidates, the Bonino list and Friends of Grillo, which together obtained 29,000 votes (1.9). In 2013, there were two which gained 160,000 preferences equally distributed between Marino (7.4) and Marchini (7.5). The same happened in 2016; Meloni’s list obtained 40,000 votes (3.4) and Marchini’s 56,000 (4.8). The use of personal lists tended to increase in the centre-left where, since 1997, they provided 638,000 votes (21.9 of the total), and where there was, perhaps, a greater tendency towards distancing themselves from the parties. The 2021 elections changed the game and all the main candidates resorted to it. Together, they reached 317,000 preferences, almost a third of the LV (31.2).²⁴ Calenda’s list almost reached 200,000 votes (19.8), almost four times more than Gualtieri’s, which topped at 55,000 (5.4).

3.4 Candidates and preferences

Between 1997 and 2021, there were 230 lists and approximately 11,300 council candidacies.²⁵ Of these, only 438 were successful (3.9).²⁶ Most of the time, these candidates were unlikely figures who entered the lists for who knows what reason, given the fact that their contribution was usually non-existent. One candidate in six (15.6) never obtained any preference at all (VP), four out of ten (41.2) did not exceed 10 and almost two thirds (65.3) did not reach 50, while only six out of a hundred (5.9) obtained over 1,000.

In the lists, the female candidates comprised just over a third (35.5), with significant changes in the last three contests, i.e., those following Law 215, which introduced double preferences and was a decisive turning point towards equal opportunities in politics (Farina and Carbone, 2016). And if between 1997 and 2013 the female candidates were less than a third (29.6), later they became more than four out of ten (43.5). The gap, however, persisted; the women elected were 101 (23.1) compared to 337 men (76.9). Changes took place following Law 215, with women elected increasing more than threefold. Between 2013 and 2021 the female candidates were 43.5 in the lists, 41.0 among those elected, and obtained 38% of the preferences. Previously (1997–2008) they were 29.6% in the lists, earned 12.9% of the total preferences and reached 12.1% among the elected candidates. If we look at the winning preferences, that is, those that led to election, before 2013 women obtained less than one in ten (8.6) and a third of the total gained by both sexes (34.9), while from 2013 onwards their weight increases in the first case to 39.8% and in the second to 45.2%.

In the four rounds preceding Law 215, preference votes (PV) were over 1, 600,000, while in the three subsequent rounds they dropped by about a third (34.4) settling at around 1 million. Nonetheless, after 2013, female preferences doubled (+92.8) and the weight of the “winning” ones increased by 150%. In addition to a greater female presence, there was also a greater probability of success. The preference vote grew significantly, there was less dispersion and, also, an increase in voter turnout, with the emergence of new female leadership and the increase of candidates capable of obtaining important packages of votes.

In the first period, 1997–2008 (P1) female candidates unable to obtain even a single preference were almost double compared to their male colleagues (21.8 against 12.2) and in three cases out of four they did not reach 50 VP (75.5) against almost two out of three among their male colleagues (60.2). Also, while men topped 500 votes in more than one round out of ten (12.5) women only managed to do so in one out of twenty (4.6). All these differences were significantly reduced in the last three rounds. Although male candidates - still at the centre of more extensive and developed political networks - were always more capable of reaching high consensus thresholds.²⁷

Female candidates had a better rate of success in the civic lists, even if this tended to decrease in the last decade. Until 2008 women candidates stood mostly for parties of the left (39.6) and centre-left (32.7), and were at their lowest for parties of the right (21.4). After 2013 they rose above the 40% threshold everywhere, although consistently maintaining a higher presence in the left-wing lists (46.5). The preferential vote followed a different dynamic from the list vote, where the votes for the parties with seats were approximately 90% of the total.

The PVs from 1997 to 2021 were 2,670,000, just over a quarter of the LVs (26.8). Those of the elected candidates were approximately 1,300,000 thousand, less than half (48.8). In the last decade, the League had the highest ratio between PV and LV (50.8), followed by the PD (43.2), An-FdI (37.2) and FI (34.7). The left-wing parties obtained on average one preference every three LVs (35.3), a little more than those on the right (32.4). At the bottom of the ranking was the M5s with one PV every ten LVs. 2013 was the year of the turning point. Previously, the civic lists obtained 11.6 of the PV, while later they reached 29.4. It should be added that personal lists, despite representing almost half of civic preferences (47.4), lost weight compared to the previous period (63.6). Considering only factions with a PV-LV ratio higher than 50%, a third were popular (32.0), just over a third were conservative centre-right (28.0) and two out of ten are progressive and leftwing (20.0) (Table 7).

Table 7

Table 7. Favoured vote and gender (1997–2021).

4 The city council

Councils were revolutionised by Law 81; their prerogatives were redefined, and they were weakened and marginalised, They also however became a hotbed of alliances, not always loyal, often at loggerheads, and jealously protecting their own local political power niche. After 1993 came some reform, but also rules confirmed and reconfirmed, even several times. City councils became launching pads for other political positions. In this last section we shall examine their components and how they changed, focusing on the voting procedures and how they can be interpreted.

4.1 The elected

Between 1993 and 2021, 499 people were elected (including those who resigned and those who replaced them) out of 396 available seats. Men were almost eight out of ten (77.8) in the 30-year period, but fewer than six (58.4) after Law 215, which allowed female councillors to go from 12.6 to 41.6. Over six out of ten women (62.2) were elected in the last decade, compared to one in four men (25.0). In 2021, they were 42.1. But it was only in 2016 that substantial gender equality was achieved (49.1 against 50.9) with the entrance of almost a quarter of the total female councillors (23.4).²⁸ Half of those elected gained their seat only once (49.9), while the other half were divided between those who did so twice (25.6) and those who did so three or more times (24.4). Those elected multiple times were 52.3 among men and 42.3 among women. Up until 2013, there were more female candidates (57.1 against 53.6); after Law 215 changed the rules this was reversed (33.3 against 48.5). Although the councillors had now greater potential to confirm their seats, it is also true that the councillors lost power when their numbers increased.

The changes took place at two moments in time: 1993 and 2013. However, let us first take a look at the makeup of the municipal assembly. Over the 30 years,²⁹ councillors belonging to a party were 83.3 and 16.7 to a civic list; over the last 10 years these figures were 75.9 and 24.1. Those belonging to a personal CL went from 11.4 to 17.5. Overall, most of them belonged within the centre-left (39.7), followed by the centre-right (18.0), the right (15.3), the populist camp (9.4), the left (8.9) and the centre (8.7). In the last three rounds, attendance dropped slightly everywhere compared to the average; not so much on the centre-left (38.6), the right (13.9), the centre (7.8) and on the left (9.0), and much more on the centre-right (6.0), which was strengthened by the entry of the populists (24.7). As regards the parties, 23.5 went to the PDS/PD, 17.1 to FI-PdL, 12.8 to MSI/An/FdI, 5.9 to Ulivo, 5.7 to RC/CI/ SEL and 4.6 to members of post-DC parties. Those elected with the M5s were 9.1. Over the last decade, over half of the seats were held by the PD (30.7), and the M5S (24.1) (Table 8).

Table 8

Table 8. Composition of the municipal assembly (1997–2021).

In the first elections held after Law 81, the new entries were, compared to 1989, three out of four (73.8) but this phenomenon soon faded away. The new political class also began to put down roots and the newly elected declined to 52.9 in 1997, to 41.7 in 2001, then up to 31.8 in 2006 and 2008. Five years later in 2013, however, the new entries comprised four out of ten councillors (41.1), in 2016 two thirds (64.2) and in 2021 almost seven out of ten (68.4). After almost disappearing after 1993, in 2013 the new entries were constantly on the rise. This change was affected by several factors that went beyond the outcome of the vote, including the introduction of new rules of the game, such as Law 81 and Law 215. Political changes such as those in the early 90s, the rise of populism in the second decade of the 2000s, and the return to the left–right axis in 2021 also played a part.

In the first round of the 90s, the replacement was general and involved all the political factions. The right, thanks to the success of the MSI, nominated various newcomers in 1993 but subsequently lost numbers. The victory of Alemanno in 2008 allowed the centre-right to introduce around 20 newly elected men into the council, more than half of the total in 30 years. The centrist lists nominated new men for the first three elections, then they also lost numbers. Even the centre-left had a higher number of replacements in the 90s, which were confirmed during the following decade. In 2008, after their defeat and the consequent reduction in seats, only “old faces” were nominated to the council. Five years later, there was another opening towards new entries, who comprised half of its councillors in 2013, half in 2016, and over two-thirds in 2021. In the last decade, the major change came with the arrival of the M5S councillors, who entered en masse in 2016 and alone account for 80 percent of the changes in the council (Table 9).

Table 9

Table 9. Elected, newly elected, gender and parties in the municipal assembly.

4.2 The municipal councillors: a profile

Our profile is based on the case of 523 individuals in office between 1986 and 2021.³⁰

4.2.1 Gender

As has already been discussed, the gender ratio did not vary substantially over the period under review. The female councillors were 14.2 before 2013 and 38.0 in the last three rounds.

4.2.2 Age

During the time frame considered, the 35–49 age group was the most numerous, with almost six out of ten councillors (57.7), while 16.8 were under 35 (16.8), 21.4 were between 50 and 64 and only 4.0 were over 65. The average age was 44.1. There were some changes over time, however. In the period from the first to the Second Republic, the number of councillors under 35 increased slightly (from 15.7 to 16.8) and rather more so of those over 40 (from 51.5 to 65.5), while those over 50 decreased (from 32.9 to 17.6). These variations are partly explained by the entry, after “Clean Hands,” of the so-called “second rankers.”. In the early 2000s, the youngest group (20.8) gained in numbers, and as the new political class became more entrenched,³¹ even the over-50s (25.9) increased, as the middle age-group (53.3) diminished. In the last decade, the 40-year-old group began to increase again in numbers (60.7), the youngest age group decreased (14.7) and, minimally, also the oldest age group (24.7). To sum up, in the transition to the 1990s, the average age dropped from 45.6 to 43.3, remained stable in the following decade (43.5) and rose slightly in the last one to 44.1.

Returning to gender, compared to the 90s, after Law 215, the youngest age group among female councillors narrowed (from 26.7 to 15.8), the median age group remained stable (from 60.0 to 59.6) and the over-fifties almost doubled (from 13.3 to 24.6). Among male councillors, after an increase in the early 2000s, the youngest group returned to previous levels (from 15.4 to 14.0), the fortyyear-olds declined (from 66.3 to 61.3) and the over-fifties increased (18.3 to 24.7). Until the 2000s, female councillors were on average a couple of years younger than their colleagues; in the early 2000s they are 4 years older, while in the last electoral rounds the ages tended to align. In the last decade, the lowest average age is found in the centre-right wing, the only group that appeared slightly rejuvenated compared to the previous period (40.2 against 41). In the centre-left it rises by 4 points (44.5) and 2 among the M5s councillors (42.7).

4.2.3 Education

In the 80s, graduates were slightly less than six out of ten (58.2), a percentage that slightly decreased in the 90s (55.5), fell below 50% in the early 2000s (42.5),³² to be mainly the majority in the following year (79.3), and close to 90% in 2021 (86.9), which was a clear improvement compared to previous decades.

At the end of the First Republic, female councillors with a degree were 45.8% versus 60.9% of their male colleagues. In the 1990s, they rose to 80% (M. 51.9), then to 71.4% in the first 2000s (M. 38.7), and to 91.2% in the last decade (M. 72.0). Compared to the 1990s, the picture was reversed in 2021, and the female component grew to 15 points above its peers (94.7 versus 79.3).

The number of graduates was higher in the centre-right (71.4) and much less in the centre-left (50.9), the left (44.0), the centre (42.9) and in the right wing (33.3). In the transition to 2000 the most skilled group dropped, to rise again everywhere in the last three elections, with values around 80% on the right and on the centre-right (81.3), the centre-left (80.4) and in the populist camp (78.9), and close to 70% on the left (71.4) and the centre (69.2) (Table 10).

Table 10

Table 10. Municipal councillors (1986–2021): age and education.

4.2.4 Socio-professional status

The sociological profile of the political class of a municipality (Barberis, 1988) depends on its size. The larger the municipality, the more the characteristics of the elected representatives tend to diverge from those of the population. Rome is the most populous city of our country as well as the capital. Throughout the entire period under consideration, the “higher” professions were the largest group (32.1) followed by the administrators (28.7), the entrepreneurs/managers (24.9) and the mid-level professions (10.9). The lower professions were very few (3.4). In the 1980s the administrator group were the first group (38.1), which decreased in the transition to the Second Republic (26.1), grew slightly at the beginning of 2000 (30.8) and decreased again in the final period (20.7). Between the 80s and the 90s the group of entrepreneurs/managers grew from 20.1 to 26.1, without significant deviations in the early 2000s (25.8) and in the last decade (27.3), although with a decline in 2021 (20.8). The higher professions - those that best represent civil society - grew in the 90s (from 30.6 to 37.8), becoming the first group. They decreased in the early 2000s (27.5) and rose again in the second decade (32.7), to become the leading group once more. However, after topping 35% in 2013 and 2016, the values fell in 2023 (22.9). Lastly, the mid-level professions were 5.2 in the 80s, then 9.2 in the 90s, 10.8 in the first half of the next decade and 17.3 in the second. Once again 2021 showed significant deviations upwards (27.1), due to a sort of interchange between the higher and mid-level professions. In the final period, rather marked gender differences emerged. Among entrepreneurs-managers, women were lower by more than 20 points (from 35.5 of men to 14.0), and 6 among administrators (from 22.6 to 17.5). The majority of the elected were from the higher professions³³ (49.1 against 22.6). In terms of affiliations, the numbers of entrepreneurs-managers increased significantly among those elected from right-wing parties (48.9), and decreased on the left (19.0) and even more among the populists (13.2). In the case of high-level professionals, there was no substantial difference between left (27.0) and right (28.9), although the majority were to be found in the populist camp (47.4). The middle-range professionals, more numerous in this last decade (23.7) are more on the left (19.0) than the right (8.9).

In the last decade, we can see some trends emerging regarding certain professions. Professional politicians and trade union leaders were always approximately one in ten (10.5 before 2013, 11.3 after). And, in contrast to the 1990s, they were increasing on the right wing (20.0). Law-based professions increased (from 7.8 to 13.3), also more on the right (15.6). The numbers of intellectual professions (academics, university professors, etc.) as well as middle school teachers (from 4.3 to 7.3) grew only slightly (from 5.0 to 7.6), both more oriented to the left wing (9.5). Journalists lost ground (from 11.8 to 5.6), and medical professions almost disappeared (from 6.7 to 0.7).

5 The executive class

Our final focus is on the executive class, that is, the leading members of the city government (council chairpersons or administrators) who arrived after the 1990s. of the 106 total, there were 76 men (69.6) and 30 women (28.3) divided between the various parties. At the end of the First Republic, most of them belonged to the Christian Democratic centre (57.9) and the lay and socialist parties (42.1). In the 1990s they all belonged to the centre-left. In the early 2000s they were split between the centre-left (60.0) and the centre-right (40.0), and in the last decade between the centreleft (61.3), the populist camp (35.5) and the left (3.2). As was the case with the councillors, the female component in the executive class has been growing continuously up to the present day. In the 80s it was 5.3, then it rose to 26.9, touched 30.0, and in the three last elections it topped 40% (41.9).

After Law 81, none of the administrators fell into the lower age bracket; almost two thirds were within the middle age group (63.2) and more than a third were over fifty (36.7), ten points above the councillors. Compared to the latter, the average age rose by almost 4 years (47.5 against 43.6) and 6 in the last decade (49.8 against 44.1). In the councils, the female group was older (48.1 against 46.7). Career transitions seemed to require a slightly longer time for women. The council executives, who on average were older, were also more trained than the council members in the 80s (89.5 against 58.2). Then, in the 90s and in the early 2000s, the level of education dropped sharply for both (71.4 against 42.5). In the last three rounds, however, there was an alignment, and the differences disappeared (80.6 against 79.3).

Female executives, who were already more trained than female councillors, were also better trained than their male colleagues (93.1 versus 65.5). In the last decade, all of them possessed a degree, which was true for only two-thirds of men (66.7). In comparison with the councillors, among the members of the executive, the higher professions were growing (48.4 against 27.5) as well as the entrepreneur-managers (32.3 against 27.3), while the mid-level professions decreased (12.9 against 17.3) and the administrative professions disappeared (3.2 against 20.7).

Over the 30 years, the first group was that of the professionals. From 1980 (10.5) they rose to 50% in the 90s, being over a third in the early 2000s (36.7) and almost half in the final decade (48.4). The entrepreneur-managers, three out of ten in the 80s (31.6), halved in the 90s (15.4), to rise in the early 2000s (23.3) and then fell back to pre-Law 81levels in the last three rounds (32.3). The administrative professionals, over half of the council executive in the First Republic (52.6),³⁴ fell in the 90s (19.2), rose again in the first years of 2000 (30.0) and were absent in the final decade (3.2). The mid-level professions, who were absent in the 1980s, had a presence between 10 and 15% over the 30 year period (Table 11).

Table 11

Table 11. Councillors and executives (1986–2021): professional groups.

6 Conclusion

As we look at Roman politics over the last 30 years, there is no shortage of themes and questions to reflect on. However, we can end with a few brief observations on the 1993 reform.

In the midst of a serious crisis, Law 81introduced a new form of city government, centred on two figures, the mayor and the citizen. The mayor became no longer a first among equals (chosen by their peers), but a first above unequals (chosen by the voters). The citizen, on the other hand, now acted as a guarantor, conferring legitimacy directly, from the ballot box. This “double act” created a new definition of governance (Caciagli, 1993), which called for leadership and personalisation, qualities which help drive voter participation (Di Franco and Gritti, 1994) and leave the party system less room for manoeuvre (Prospero, 1994). Until the early 2000s these demands were satisfied; Rutelli and Veltroni (like other mayors in other cities) enjoyed recognition, and possessed influence and charisma which were missing in those who succeeded them. Recognition led to mobilisation, following and feedback. Personalisation was its realisation. The mayor became the “First Citizen” legitimised by the recognition of him as such from below,³⁵ from a base that was broad and extensive. Between 1993 and 2008, participation in elections was on average, about 74%. In the last three elections it dropped to 53%. One of the two figures seemed to want to abandon the field, especially in the last election, with 48% in the first round and 40% in the run-off, amounting to a veritable stampede; and as one figure, the elector, left the field, the other, the mayor, began also to lose ground.

In the first two decades after Law 81, the average vote for the mayor was 47%. In the last 10 years, it dropped to 34%, losing a quarter of the consensus. In the first three rounds, mayoral votes exceeded those of the mayoral list by 35%, and in the last three by 12%, almost flattening out in 2021 (+6%). The electoral field was narrowing; from an electoral base of a third of those entitled to vote (35%) numbers decreased dramatically to 18%. In 1997, Rutelli enjoyed the support of 43% of the electoral base, while Gualtieri never reached 13%. Which were hardly ideal conditions for leadership and personalization. These shortcomings soon became the subject of debate, in the administration, in the interaction with other political parties, in the operation of governance. It was projected towards the outside, to civil society and the citizens themselves.

Leadership needs participation, of a group (a team that becomes a party) where it can operate and draw support and has an audience in the public theatre (Haslam et al., 2013). This is the mayor of the “era of mayors.” Then consensus and participation fade away and the audience no longer attends the performance. The “first citizen” is reduced to the “person nominated.” It is the end, not of the “era of mayors,” which has already finished, but of a political idea that has lost its interpreters. Born in an anti-political climate, it ends up extinguishing itself, paradoxically, with the advent of populism and, therefore, with a disorderly and weak reappearance of the parties, with the arrival of the fluid vote and non-vote.

Today the idea of participation and representation needs to be rethought, starting from the idea of a minority legitimising base. A “first citizen” mayor chosen by one voter in ten is not in keeping with the spirit of Law 81, which looked towards “the citizens’ mayor” and the “citizens who choose.”

Voting expresses adhesion, support and approval (Mannheimer and Sani, 1987), so in non-marginal abstention, deficits of legitimacy and weakening of trust emerge. This is a question of mechanisms, substantiality and the substance of representation, and therefore, of democracy. Voting is not a duty, but a civic right and “abstentionism is a way of voting” (Pasquino, 2014, pp. 241–242). It is detachment, disinterest and apathy that explain estrangement and alienation, and of course it should be a reasoned choice, an active reaction that expresses protest, criticism and dissociation (Mannheimer and Sani, 2001).

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Author contributions

TF: Writing – original draft, Writing – review & editing.

Funding

The author declares that financial support was received for the research and/or publication of this article. The publication of this article was funded by the Department of Economics, Society and Politics (DESP) at the University of Urbino.

Conflict of interest

The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The reviewer DS declared a shared affiliation with the author(s) to the handling editor at the time of review.

The handling editor RDS declared a shared specialised society (https://atplc.org/organizzazione/) with the author(s) at the time of review.

Generative AI statement

The author declares that no Gen AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

^Under 15,000 people mayors are elected in the first round (plurality) and the list associated with them obtains two thirds of the seats in the council.

^Between 1993 and 2021, EVs grew by 42 thousand units (+1.8).

^Values are reported to the nearest thousand.

^With the exception of 2001, the VS involves approximately 95% of voters, with a slight increase in the last electoral rounds.

^Due primarily to the drop in turnout.

^Within the new system the competition is limited to the two main actors and all the others become supporting actors (Mancini, 1997).

^The storey must be read with reference to Rutelli’s candidacy (without primaries) punished by the voters in the run-off (see Cerrina, 2020).

^38 lists and 22 mayor candidates.

^The first seat for the list or coalition supporting the candidate is automatically awarded to the candidate.

^Previously it happened only to Alemanno in 2008.

^Among the second place finishers, Giacchetti in 2016 received the least votes, despite having 25 thousand more votes than Gualtieri.

^The personal vote exceeds that of the lists even in the case of the second-placed candidates. Fini in 1993 goes beyond by 43.9, while Michetti in 2021 manages to do so by only a few thousand votes (5.0).

^The vote for the mayor corresponds, instead, to 74.6 of those entitled to vote in 1993 and 47.0 in 2021.

^The average is 31.8. 28.6 between 1993 and 2008 and 37.3 afterwards (2013–2021).

^The lists below 1% are 47% in the 90s and two thirds later.

^Except in 2016 when the excluded lists reached almost 20% of the VL.

^Parties established in the local and national electoral market and as part of defined political families.

^The votes of Msi-An-FdI are around one million and 600 thousand. Those of 2021 are 11% of the total.

^FI and Pdl (1997–2021) get about 1.3 million votes. Votes in 2021 are less than 3%.

^A package that corresponds to 51% of the VL in 2021.

^The sum of the Pds, Ulivo and Pd votes is approximately 2 million and 350 thousand. In 2021, 7% of the total.

^In 1989 the PSI obtained 246 thousand preferences (13.7).

^Votes for the Right-Tricolour Flame which proposes Storace as mayoral candidate.

^The contribution of personal lists on VL is 15.9%.

^MV 1,600 per election. In 1997 below 1,300, in 2006 they almost reached 2000 and in 2021 about 1,650. In 18.8 (2,124 cases) they were nominated on more than one occasion and in 6.1 (689) on more than two.

^Among the elected there are also those who resigned and those who took over.

^Before 2013, the female candidates with above 1,000 preferences were 2.7 against 8.0 of males. Subsequently, they succeeded in 4.7 of cases, compared to 5.8 among their male colleagues. Excluding Fini and D’Alema,—unreachable with 63,000 and 36,000 preferences in 1997—in the first ten positions for PV there were four female candidates (Belviso, Marino, Mussolini and Alfonsi), all elected after Law 215.

^2021 (21.6) and 2013 (17.1). The lowest female rate is recorded in 2008 (6.1).

^The period considered is 1997–2021.

^Ministry of the Interior (https://dait.interno.gov.it/elezioni/anagrafe-amministratori).

^During this period, confirmations reach higher levels.

^The municipal council with the fewest graduates was that of 2008 (35.0).

^Those professions were clearly male dominated until the 1990s.

^During the years of the First Republic, membership in the bureaucracy was often a common denominator in municipal councils (Barberis, 1988).

^In Venice Cacciari dusts off the figure of the Doge (Fava, 2002).

References

Agosta, A. (1999). “Sistema elettorale e governo locale: gli effetti della riforma del 1993” in Votare in Città, (a cura di). ed. S. Operto (Milano: Franco Angeli).

Google Scholar

Baldini, G.Pappalardo., (2004), Sistemi elettorali e partiti nelle democrazie contemporanee. Roma-Bari: Laterza.

Google Scholar

Baldini, G., Bucchi, M., and Fava, T. (2000). “Comunicazione politica e decisione di voto: una ricerca sulle elezioni comunali del 1999” in Mass Media ed elezioni. ed. G. Sani (a cura di) (Bologna: Il Mulino) pp. 225–274.

Google Scholar

Barberis, C. (1988), La classe politica municipale. Milano: Angeli.

Google Scholar

Bettin Lattes, G., and Magnier, A. (1997). I nuovi sindaci. Come cambia una carriera politica. Riv. Ital. Sci. Polit. 1, 91–118.

Google Scholar

Burroni, L., Piselli, F., Ramella, F., and Trigilia, C. (2009). Città metropolitane e politiche urbane. Firenze: Firenze University Press.

Google Scholar

Caciagli, M. (1993), Conclusioni, in G. Riccamboni (a cura), Cittadini e rappresentanza in Europa, FrancoAngeli, Milano.

Google Scholar

Camarda, L. (2018). Il consigliere comunale. Milano: Key Editore.

Google Scholar

Canzano, A. (2007). Dall’enclave bianca all’iperpopulismo. Soveria Mammeli: Rubbettino.

Google Scholar

Canzano, A. (2012). Sindaci e giovani sindaci. Le dinamiche di trasformazione della politica locale. Milano: FrancoAngeli.

Google Scholar

Carrai, M. (1997). Nuove regole elettorali e subculture politiche. Quad. dell’Osserv. Elettor. 37. pp. 43–92

Google Scholar

Catanzaro, R., Piselli, F., Ramella, F., and Trigilia, C. (2002). I comuni nuovi. Bologna: Il Mulino.

Google Scholar

Cazzola, F. (1991). Periferici e integrati. Chi, dove e quando nelle amministrazioni comunali. Bologna: Il Mulino.

Google Scholar

Cedroni, L., and Fava, T. (2013). Roma. La periferia al Centro, in Fava …, 129–148.

Google Scholar

Cerrina, A. 2020 Le elezioni comunali e la politica a Roma dal 2008 ad oggi, Tesi di laurea, 2019–20, Luiss, Roma.

Google Scholar

della Porta, D. (1999). La politica locale. Bologna: Il Mulino.

Google Scholar

Diamanti, I. (2003a). Bianco, rosso, verde, azzurro. Mappe e colori dell’Italia politica. Bologna: Il Mulino.

Google Scholar

Diamanti, I. (2003b). Così l’era dei sindaci giunge al tramonto. La Repubblica. 26 maggio

Google Scholar

Diamanti, I. (2008). Mappe dell’Italia politica Bianco, rosso, verde, azzurro … e tricolore. Bologna: Il Mulino.

Google Scholar

Di Franco, G., and Gritti, R. (1994). L’Italia al voto. Roma: Edizioni associate.a cura di

Google Scholar

Di Virgilio, A. 2006 Il sindaco elettivo: un decennio di esperienze in Italia M. Caciagli and A. VirgilioDi Eleggere il sindaco. La nuova democrazia locale in Italia e in Europa Torino: Utet, 5–24.

Google Scholar

Fabbrini, S. (2001). Features and implications of semi-parliamentarianism: the direct election of Italian mayors. South Eur. Polit. Soc. 2, 47–70. doi: 10.1080/714004942

Crossref Full Text | Google Scholar

Farina, F., and Carbone, D. (2016), Oltre le quote: sguardi plurimi delle elette. Indagine nazionale sulle consigliere comunali. Sant’Arcangelo di Romagna (Rn). Maggioli editore.

Google Scholar

Fava, T. (2002). “Venezia” in Catanzaro e altri…, 464–498.

Google Scholar

Fava, T. (2013). Il Partito democratico visto dal basso. Roma: Aracne.

Google Scholar

Fava, T. (2022). La destra e la sinistra nell’era dei populismi. Uno studio di caso. Polis XXXVI, 219–256.

Google Scholar

Gaspari, O., Forlenza, R., and Cruciani, S. (2009). Storie di sindaci per la storia d’Italia. Roma: Donzelli.

Google Scholar

Giglioli, P. P., Cavicchioli, S., and Fele, G. (1997). Rituali di degradazione: anatomia del processo Cusani. Bologna: Il Mulino.

Google Scholar

Haslam, S. A., Reicher, S. D., and Platow, M. J. (2013). Psicologia del leader. Bologna: Il Mulino.

Google Scholar

Ignazi, P. (2014). Vent’anni dopo. La parabola del berlusconismo. Bologna: Il Mulino.

Google Scholar

Inglehart, R., and Klingermsnn, H. D. (1976). “Party identification, ideological preference and the LeftRight dimension among Western mass publics, in (a cura di)” in Party identification and beyond: Representations of voting and party competition. eds. I. Budge, I. Crewe, and D. Farlie (London: Wiley), 243–273.

Google Scholar

Lombardo, C. (1984). “Alla ricerca del centro perduto” in L’Italia al voto …. eds. G. Di Franco and R. Gritti, 79–98.

Google Scholar

Mancini, P. (1997). Candidati sindaci in prestito dalla politica. Comun. Polit. VI:7–9.

Google Scholar

Mannheimer, R., and Sani, G. (1987). Il mercato elettorale. Identikit dell’elettore italiano. Bologna: Il Mulino.

Google Scholar

Mannheimer, R., and Sani, G. (2001). La conquista degli astenuti. Bologna: Il Mulino.

Google Scholar

March, J. G., and Olsen, J. P. (1992). Riscoprire le istituzioni. Bologna: Il Mulino.

Google Scholar

Pasquino, G. (2014). Partiti, istituzioni, democrazie. Bologna: Il Mulino.

Google Scholar

Piselli, F., Burroni, L., and Ramella, F. (2012). Governare la città. Beni collettivi e politiche metropolitane. Roma: Donzelli.

Google Scholar

Prospero, M. (1994), “Il presidenzialismo comunale: l’elezione diretta del sindaco” in L’Italia al vot o …. eds. G. Di Franco and R. Gritti, 9–14.

Google Scholar

Revelli, M. (2017). Populismo 2.0. Torino: Einaudi.

Google Scholar

Revelli, M. (2019). Politica senza politica. Torino: Einaudi.

Google Scholar

Sartori, G. (1995), Ingegneria costituzionale comparata. Bologna: Il Mulino.

Google Scholar

Schadee, H., Segatti, P., and Vezzoni, C. (2019). L’apocalisse della democrazia italiana. Alle origini di due terremoti elettorali. Bologna: Il Mulino.

Google Scholar

Segatori, R. (2003). I sindaci. Storia e sociologia dell’amministrazione locale in Italia dall’Unità a oggi. Roma: Donzelli.

Google Scholar

Tarrow, S. (1979). Tra centro e periferia. Bologna: Il Mulino.

Google Scholar

Vaciago, G. (1999), Il sindaco uno e trino, “Il Mulino”, 1.

Google Scholar

Vandelli, L. (1997). Sindaci e miti. Bologna: Il Mulino.

Google Scholar