Faraway, so close The formation of the Conte government in Italy, 2018
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Faraway, so close
The formation of the Conte government in Italy, 2018∗
Daniela Giannetti†1 , Andrea Pedrazzani‡1 , and Luca Pinto§1
1
Department of Political and Social Sciences,
University of Bologna
August 17, 2018
Paper prepared to be presented at the
ECPR General Conference 2018, Hamburg, 22-25 August
Abstract
The “yellow-green” government led by Giuseppe Conte that formed in Italy
after the general election held on March 4 2018 is composed by two parties
that share a generic anti-elite populist rhetoric but appear quite distant on
the left-right continuum. This paper aims to cast light on such outcome which
seems to represent a deviant case from the main theories on coalition formation
developed in the literature. In order to do so, we combine a statistical anal-
ysis of the drivers of government formation in Italy between 2001-2018 with
an in-depth study of the structure of the policy space characterising party
competition in 2018. Using data from an expert survey fielded by the authors,
we found that political competition can be better analysed on the basis of a
two-dimensional spatial framework. The most important dimension is related
to immigration, EU issues and social conservatism. The second one coincides
with the economic left-right. The outcome of the government formation pro-
cess appears more understandable once parties’ policy positions are measured
along these two dimensions, rather than on the general left-right continuum.
Keywords: policy space; expert survey; populism; government formation; coalition
government; Italy
∗
This work was supported by the Italian Ministry for Education, University and Research,
with the grant [2015P7RCL5], “Politics and Policy in Europe in times of crisis: Causes and conse-
quences”. This is a preliminary draft. Please, do not cite or distribute without permission of the
authors.
†
Email: daniela.giannetti@unibo.it.
‡
Email: andrea.pedrazzani@unibo.it.
§
Email: luca.pinto@unibo.it.
11 Introduction
Coalition formation theories predict that governments are more likely to form when they
satisfy a combination of requirements: they should control a majority of seats in the par-
liament – possibly a minimal winning majority that do not contain any unnecessary party
– and should include partners with similar ideological backgrounds (Laver and Schofield
1990; Laver 1998; Martin and Stevenson 2001). In this way, governing parties are expected
to win control over as many cabinet portfolios as possible and implement policies that are
the closest to their ideal policy preferences. These predictions are driven by the assump-
tion that parties are rational actors that pursue three major goals: they want to maximize
their utility expressed in office payoffs, their influence on policy proposals and their share
of votes in any election, as gaining office and controlling policies are only possible through
winning the elections (Müller and Strøm 1999). However, more than often governments
that actually form in the real world do not follow such predicted pattern.1 At first glance,
the new Italian government that was created in the aftermath of the general election held
on March 4, 2018 – which is composed by two parties that apparently have little or nothing
to share other than a generic anti-elite populist rhetoric – is a perfect example of a deviant
case.
At the end of May 2018, after about 70 days of thrilling negotiations, the League (former
known as Northern League, LN) and the Five Star Movement (M5S) reached an agreement
to form a government headed by Giuseppe Conte, a lawyer and university professor close
to the M5S, but without any significant previous political experience. These two parties,
which taken together won more than 50 per cent of the votes cast, contested the election
proposing two very different and – according to many observers – incompatible policy
platforms. The LN, which won support especially in the northern part of the country,
offered a policy package which included lower taxes on personal income and protectionist
measures against global trade. On the contrary, the M5S, whose success was achieved
thanks to the political support in the southern part of Italy, promised an increase in
public spending and the start of a new program that would award generous benefits to the
unemployed and to low-income workers (Anelli et al. 2018). These differences regarding
economic policy, together with additional disagreements concerning other issues such as
environmental protection, seem to push the two governing parties on opposite sides of
the overarching socio-economic left-right spectrum which, according to most studies of
coalition formation, constitutes the most salient dimension shaping political competition
1
Martin and Stevenson (2001) concluded that about 40 per cent of the governments that actually
formed in Western Europe are predicted by their model. In replicating the model of Martin and
Stevenson on a different selection of cases, Bäck and Dumont (2007) found a predictive rate of 32
per cent.
2in Europe (Benoit and Laver 2006).2
The literature on coalition formation analysed deviations between theoretical predic-
tions and real outcomes through in-depth case studies (Bäck and Dumont 2007). This
paper follows the same strategy attempting to cast light on the formation of the Conte
government, which seems to represent a relevant deviant case. Given the very different
policy platforms of the LN and the M5S, how can we explain the formation of a govern-
ment including these two parties? To solve the puzzle, we combine statistical analysis
with an in-depth study of the structure of the policy space characterising party competi-
tion in the 2018 election. We use first a conditional logistic regression model to identify
the factors that drove government formation in Italy since 2001. Such approach has been
widely adopted in coalition studies since it was introduced in the literature by Martin and
Stevenson (2001).3 The analysis reveals that our model fails to correctly predict the actual
government that formed after the 2018 election. Then, we investigate the reasons of this
failure by exploring the structure of party competition in Italy during the election of March
2018. Relying on the spatial approach to party competition to analyse the most salient
dimensions of the policy space in the Italian context, we use data from an original expert
survey fielded by the authors to show that the general left-right dimension used in the
statistical model to measure parties’ ideological compatibility is not adequate to correctly
predict party competition.
Our analysis highlights a two-dimensional policy space configuration defined in the
first place by a super-issue related to immigration, attitudes towards the European Union
(EU) and social conservatism, and in the second place by a dimension related to the
economic left-right. This configuration, which highlights the salience of a “demarcation-
integration” dimension (Kriesi et al. 2006, 2008), represents a further evidence of the
changing shape and structure of the Italian policy space following the Great Recession
(Giannetti, Pedrazzani and Pinto 2017).The outcome of the government formation process
after the election of March 2018 results more understandable once parties’ policy positions
are measured along these two dimensions, rather than on the classic left-right continuum.
By focusing on the Conte government case, this study contributes to evaluate how
general patterns of government formation in Italy have changed due to changing axes of
2
Hand coded textual analysis of party manifestos show policy divergences between M5S and
the League (Valbruzzi 2018). However, a comprehensive analysis of the reliability and validity of
different estimates of Italian parties’ policy positions, and how they correlate, is still lacking.
3
This period encompasses five legislatures and eight government bargaining processes, excluding
the one that led to the formation of the caretaker cabinet led by Mario Monti (2011-2012). It
covers most of the so called “Italian Second Republic”, in which a series of reforms – starting
from the electoral one – fundamentally altered the model of party competition, promoting the
transformation from a consensual political system (typical of the First republic, 1946-1993) to a
more majoritarian one, characterised by new political actors, bipolar competition and frequent
alternation in government (Cotta and Verzichelli 2007). Given the major differences between the
two periods characterising Italian political history, we choose to focus on the latter one in order to
better isolate the effect of our variables. For an analysis of government formation during the First
republic see: Curini and Pinto (2013, 2017).
3party competition and the rise of anti-establishment parties in the last decade. Moreover,
by combining statistical analysis with a case study, this work reaffirms the importance
of a combined research strategy to improve our understanding of coalition governments
formation. Finally, our analysis enables us to develop a better knowledge of within-country
variations in cabinet formation in a way that it is often difficult to achieve in large cross-
national studies.
The paper proceeds as follows. In the next section, we will briefly review the main
coalition theories presented in the literature. In the second section we will present the
results of a model of government formation and the data and methods used to develop it.
Next, we assess the predictive performance of our statistical analysis. In the third section,
we present a brief account of the Italian election of March 2018 and of the Conte government
formation. In the fourth section, we first overview the expert survey methodology. This
method is then used to explore the dimensionality of the Italian policy space. Finally, we
present a two-dimensional map of the Italian party system. Concluding remarks follow in
the final section.
2 Government formation in Italy, 2001-2018
2.1 Theories of coalition formation
Theories of coalition formation can be broadly divided into three set of models: the office-
seeking, the policy-seeking and the neo-institutionalist ones (for a review see: Laver and
Schofield 1990; Laver 1998; Martin and Stevenson 2001). The office-seeking theories elab-
orated a series of propositions about government formation in multiparty systems mainly
concerning coalitions’ size. The “winning” proposition states that only majority cabinets
will form, stressing the idea that majority status is a core feature of parliamentary gov-
ernment. This means that potential governments are more likely to form if they control a
majority of seats in the parliament. The “minimal winning coalition” proposition refines
the winning proposal predicting that only coalitions that do not contain any unnecessary
partner for reaching a majority will form (von Neumann and Morgenstern 1944; Riker
1962). Hence, potential governments are more likely to form if they are minimal winning
coalitions. However, given the wide set of outcomes predicted by the previous propositions,
scholars proposed more restrictive versions of the minimal winning solution. One of these is
the “bargaining proposition”, according to which parties will prefer coalitions including the
smallest number of parties in order to reduce transaction costs (Leiserson 1966, 1968). It
follows that potential governments are more likely to form the fewer the number of parties
they contain.
The predictions derived from office-seeking models were often contradicted by empiri-
cal evidence, as many real-world coalitions clearly violate the propositions discussed above
(for the prediction rates of the various set of theories see: Bäck and Dumont 2007; Martin
4and Stevenson 2001). Considering this, scholars suggested an alternative set of models,
assuming that parties are not primarily driven by office-seeking motivations, but by policy
concerns. Policy-seeking theories expect that parties with a similar ideological background
should form stable coalitions. Consequently, they predict coalitions consisting of adjacent
parties or that have the smallest programmatic distance on the left-right continuum (De
Swaan 1973; Axelrod 1970) which, according to most coalition studies, shape party com-
petition in Europe (Benoit and Laver 2006). As a consequence, potential governments are
more likely to form the lower is their ideological range.4
Starting from the 1980s, the neo-institutionalist approach began to emerge as the major
one complementing office- and policy-seeking theories. New-institutionalist scholars put the
emphasis on the role of different types of norms and institutions structuring the outcome
of the coalition formation processes (Martin and Stevenson 2010; Strøm, Budge and Laver
1994). Among these, pre-electoral commitments to govern together have proven to be an
important predictor of the coalitions that actually form. Due to the importance given by
parties and voters to credibility, these statements – which are usually public – constitute a
powerful restriction to coalition bargaining (Debus 2009; Golder 2006). We should therefore
expect that potential governments are more likely to form if they are based on pre-electoral
coalitions. Similarly, another element affecting coalition bargaining is the existence of
a status quo coalition which, under certain conditions, can enjoy a privileged position
in the government formation game (Laver and Shepsle 1996). Consequently, potential
governments should be more likely to form the more they mirror the incumbent coalition.
2.2 Data, analysis and results
To identify the factors that affect government formation in Italy, we follow Martin and
Stevenson (2001), assuming that each government negotiation represents a formation op-
portunity, in which parties choose a coalition among many possibilities. The number of
potential governments that could form depends on the number of parties represented in
the parliament. In general, for p parties, 2p - 1 potential coalitions exist. This leads to
a dataset including 8,120 potential governments that could have formed in Italy between
2001 and 2018, grouped in eight formation opportunities.5
For each potential coalition in the dataset, we identify a series of attributes based on
4
A further series of hypotheses in government formation studies focused on the identity of
specific actors that are more likely to enter the government. Austen-Smith and Banks (1988)
stress the prominent position of the largest parliamentary party in the coalition game, which
cannot be easily excluded from winning coalitions. Policy-based theories suggest instead that
when parties compete on a single ideological dimension, the party controlling the median legislator
position should have a privileged position in coalition negotiations (Black 1958; Downs 1957). In
the period under observation in Italy all the governments that actually formed included the largest
or the median party on the left-right dimension.
5
We discard the formation opportunity associated with the negotiation over the caretaker gov-
ernment headed by Mario Monti because it was formed exclusively by technical ministers without
any official party affiliation.
5office- and policy-based incentives and institutional constraints. Using information on the
distribution of seats in the lower and upper chamber composing the Italian parliament, we
create two dummy variables, coding as one all the alternatives that control a majority of
seats both in the Chamber and the Senate and those that are minimal winning coalitions
(in the lower chamber).6 Information about the absolute number of parties included in
each potential coalition are used to test the bargaining proposition. To test policy-based
features of potential coalitions, we compute for each of them the ideological range as the
distance between the left-most and the right-most bargaining party on the general left-right
scale. As a source of data for parties’ policy positions, we use five surveys administered to
Italian experts between 2001 and 2018 following the methodology developed by Laver and
Hunt (1992) and Benoit and Laver (2006). Finally, to assess the impact of institutional
constraints in structuring government formation, we include two further variables. The
first one is a dummy identifying all the potential governments that are based on a pre-
electoral coalition; the second one is a continuous measure of the proportion of parties
represented in each potential coalition that were also represented in the previous cabinet
(returnability).7
To test how the potential coalitions’ attributes listed above affect the likelihood of
government formation in Italy, we rely on a conditional logistic regression. Through this
technique, government formation can be modelled as an unordered discrete choice problem
where each formation opportunity represents a case, while the set of alternatives is struc-
tured by all the potential coalitions of parties that might form the government. Among all
these potential coalitions, only one will correspond to the real government. The results are
reported in Figure 1, which plots unstandardised conditional logit coefficients and their 95
per cent confidence intervals for two models: the first one (M1) includes all the potential
coalitions; the second one (M2) imposes a majority constraint, selecting in each forma-
tion opportunity only those alternatives that hold a majority in the lower chamber.8 The
confidence intervals inform us if a covariate increases or decreases in a significant way the
likelihood that a potential government will form. When the confidence intervals are both
6
Maintaining that negotiators aim to control both branches of a bicameral parliament, we may
assume that they will try to assemble a legislative majority not only in the lower house, but also
in the upper house, especially in a country such as Italy in which the government needs to pass
an investiture vote in both chambers. This should “double” all majority considerations that enter
the process of coalition formation (Müller, Bergman and Strøm 2008, p. 24). Achieving a double
majority is more difficult when the two chambers register increasing levels of incongruence, such
as those observed in Italy since 2001 (Pedrazzani 2017). Although, the second chamber plays a
direct role in government formation, the lower house constitutes the main arena of negotiations.
For this reason, the existence of minimal winning coalitions is investigated only in relation to the
distribution of seats in the lower chamber.
7
During the 2018 elections, Italian citizens cast their ballot for the parliament under a new
electoral system. This is the fourth system adopted in Italy since 1994 (Baldini 2011; Chiaramonte
2015). Besides the differences, all of them create incentives for parties to form pre-electoral alliances
in order to compete in the elections.
8
All the governments that actually formed in Italy in the period under observation hold a
majority in the lower chamber.
6M1 M2
Majority government (in both chambers)
Minimal winning coalition
Number of parties
Variable
Ideological range (left-right)
Pre-electoral coalition
Returnability
-5 0 5 -5 0 5
Coefficient
Note: Parameter estimates are unstandardized conditional logit coefficients with 95 per cent con-
fidence intervals. M1: Observations (potential governments): 8,120, grouped in eight formation
opportunities. Pseudo-R2: 0.41; AIC (Akaike Information Criterion): 70.99; BIC (Bayesian Infor-
mation Criterion): 113.01. M2: Observations: 3,911, grouped in eight formation opportunities;
Pseudo-R2: 0.49; AIC: 57.16; BIC: 94.79.
Figure 1: Determinants of government formation, Italy 2001-2018
on the right (or on the left) of the zero line, coalitions’ attributes positively (or negatively)
influence in a statistically significant way the likelihood of government formation.
The first model shows that most of the coefficients have the expected effects. Potential
governments which hold a majority status in both chambers and are minimal winning
coalitions have a higher chance to form. Similarly, potential coalitions are more likely to
form if they are based on a pre-electoral coalition. Conversely, the greater the distance
between the left-most and the right-most party in a government alternative, the lower
its likelihood to become the actual government. Finally, the number of parties and the
proportion of incumbent parties in potential coalitions do not exert any significant impact
on government formation. When exclusively majority alternatives are included in the
analysis, only two variables result to be significant: ideological range and pre-electoral
coalitions. This means that potential majority governments are more likely to form if they
constitute a pre-electoral alliance and minimise the ideological divisions among partners.
72.3 Predictive performance of government formation models
To evaluate the predictive performance of coalition theories on the Italian case, we gener-
ate predicted probabilities from the models presented above, assuming that the potential
government with the highest predicted probability is the one that will form (Bäck and
Dumont 2007; Martin and Stevenson 2001). If the models correctly predict the outcome of
the coalition formation game, the difference between the predicted probability of the actual
government and of the potential government with the highest predicted probability (i.e.
the predicted government) will be equal to zero. Otherwise, we should obtain a difference
higher than zero. Figure 2 illustrates these deviations for the eight formation opportunities
included in our analysis.
Figure 2 shows that, according to M1, the predicted government corresponds to the
one that actually formed for three out of eight formation opportunities – Berlusconi II,
Berlusconi III and Berlusconi IV, for which the difference in the predicted choice probability
is zero. This means that the prediction rate of M1 – the number of predicted outcomes
divided by the total number of formation opportunities included in the analysis – is about
38 per cent, which is a figure similar to that reported by other well-known cross-national
studies (Martin and Stevenson 2001). This figure rises to 50 per cent considering M2,
which is able to predict correctly also the Prodi II government. These prediction rates
can be considered quite remarkable if we think that real governments are picked up from
thousands of potential alternatives. However, it should be noted that they have been
obtained by combining several variables related to different theories, among which those
highlighted within the neo-institutional approach, in particular the role of pre-electoral
coalitions, appear prominent.
A careful investigation of the choice probabilities produced by the first model reveals
that predicted governments – at least in post-election situations – tend to correspond to
the pre-electoral coalition which minimises the ideological distance between partners along
the left-right dimension, no matter if it constitutes the majority or not. When pre-electoral
alliances are not ideologically compact or do not hold a majority both in the lower and
upper houses – such as the heterogeneous coalition led by Romano Prodi (2006) which was
supported externally in the Senate by small parties not directly involved in the government
– our model fails to produce a correct prediction. On the contrary, when a pre-electoral
alliance wins a majority in both chambers, the first model produces correct predictions,
as in the case of the three cabinets led by Silvio Berlusconi (2001, 2005, 2008). When
imposing the majority constraint, as in the second model, predicted governments usually
correspond to the winning pre-electoral coalition resulting from the elections. When none
of the pre-electoral alliances obtains a majority of seats – as in the case of the Letta (2013)
and Conte (2018) governments – the model picks up the combination of parties which
minimises the left-right ideological range.
The worsening in predictive performance of our models we observe starting from the
8Difference in predicted choice probability 1.00
0.75
0.50
0.25
0.00
14) a
16) i
05) II
06) I
08) II
201 i IV
18) i
18- te
14- Renz
16- ilon
PE
IE
PE
E
PE
IE
IE
E
05- i II
13- Lett
01- oni
06- odi
(20 Con
1) P
)P
(20 uscon
08- con
(20 Gent
(20 rlusc
(20 Pr
(20 erlus
rl
Be
(20
Be
(20
B
Government
Model M1 M2
Note: The Y-axis shows the difference between the predicted probability of the actual govern-
ment and the government with the highest predicted probability (i.e. the predicted government)
according to the estimates of the conditional logit models presented in the paper. The difference
is plotted for each formation opportunity included in our analysis. PE stands for “post-electoral”
governments; IE for “inter-electoral”. Governments are shown in chronological order.
Figure 2: Predictive performance of government formation models, Italy 2001-2018
Letta government coincides with the success of the M5S, which represents one of the most
significant occurrences in the recent Italian political history due to its impact on the party
system (Tronconi 2015a, 2018). In 2013 the M5S, participating in its first ever national
election, became the most voted party, breaking the influence of the centre-left and centre-
right pre-electoral coalitions, which had dominated electoral competition since 1994. The
emergence of the M5S also determined a redefinition of the policy space, contributing
to the politicising of the EU related issues (Di Virgilio et al. 2015). While before the
2013 EU-related issues were associated with economic, immigration and social policies in
a super-issue that may be interpreted as a general left-right, since 2013 they have become
more distinctly aligned with a new dimension of political competition tapping pro-/anti-EU
attitudes (Giannetti, Pinto and Pedrazzani 2017).
Our analysis reveals therefore that the incapacity of the models presented above to
correctly predict the governments that actually formed in Italy starting from 2013 derives
essentially from two interrelated factors. The first one is related to a highly volatile electoral
environment, which resulted in the change of the structure of party competition from
bipolar to tripolar (Chiaramonte and Emanuele 2017; Tronconi 2015b). The second factor
9is the inappropriateness of the general left-right dimension to accurately predict patterns
of political competition. The first factor can be easily controlled by excluding from the
analysis all the combinations of parties that do not constitute a majority in the parliament:
in this way minority pre-electoral coalitions are dropped, and the formation game is limited
to those alternatives that can pass an investiture vote in the parliament. Our results
indicate that among these potential governments the one which minimises the left-right
ideologically range tends to be chosen: if this prediction consistently deviates from the
one associated to the real government, then we should explore if the general left-right
dimension still represents an adequate framework to describe parties’ ideological positions
in Italy after the Great recession.
3 The Italian general election of March 2018 and the Conte government
On March 4, 2018 Italy voted to renew the Chamber of Deputies and the Senate. Italian
citizens cast their ballot under a new electoral system: a mixed system in which approxi-
mately one third of the seats are allocated in single-member districts using plurality rule,
while two thirds are assigned through a proportional formula. Despite the presence of a
plurality tier, the effects of the new electoral rules on the allocation of seats between parties
were by and large proportional (Chiaramonte and D’Alimonte 2018).
According to the electoral system, votes are allocated to two types of collective political
entities: pre-electoral coalitions (PECs) and parties within them. In March 2018, only two
coalitions consisted of more than one party: the centre-right (CR) and the centre-left (CL)
electoral alliances. The centre-left coalition was led by the Democratic Party (PD), which
was the main partner of the incumbent government coalition, and the party of the former
prime minister, Paolo Gentiloni. The PD contested the 2018 elections in alliance with
More Europe (+EU) and other minor lists, which failed to pass the threshold imposed by
the electoral system to be represented in the parliament. The centre-right coalition was
made up of Go Italy (FI), the League (LN), Brothers of Italy (FdI), and the minor list
We are with Italy (NcI). The other coalitions that contested the 2018 election actually
consisted of only one party. The main ones were the following: the Five Star Movement
(M5S), Free and Equal (LeU), and Power to the People (PaP).
Table 1 reports the electoral results for the two branches of the Italian parliament,
together with the number of seats won by each party and their ideological position on the
general left-right scale employed in the statistical model presented in the previous section.
Party placements are derived from an expert survey fielded by the authors in the aftermath
of the election, based on the format developed by Laver and Hunt (1992) and Benoit and
Laver (2006) (see below for more information on the survey). Table 1 shows that none of
the two pre-electoral alliances was able to reach a majority in the parliament, unless allying
with the M5S or forming a “grand coalition” between the two electoral cartels. Left-right
party placements reveal that LN and FdI are the two most far-right parties. PaP and LeU
10Table 1: Results of the Italian general election of March 2018
Vote % Vote % Seat % Seats %
PEC Party Left-right
(Chamber) (Senate) (Chamber) (Senate)
CR LN 17.350 17.610 0.198 0.181 18.324
CR FI 14.000 14.430 0.165 0.191 15.268
CR FdI 4.350 4.260 0.051 0.056 18.380
CR NcI 1.300 1.200 0.006 0.000 12.943
CL PD 18.760 19.140 0.176 0.163 7.986
CL +EU 2.560 2.370 0.005 0.003 10.141
– M5S 32.680 32.220 0.352 0.341 12.329
– LeU 3.390 3.280 0.022 0.013 4.314
– PaP 1.130 1.060 0.000 0.000 3.268
Note: The general left-right dimension ranges from 1 (Left) to 20 (Right). Experts
are asked to locate parties, taking into consideration all aspects of party policy into
account. CR: centre-right pre-electoral coalition; CL: centre-left.
Source: Ministry of Interior (https://elezionistorico.interno.gov.it/.)
are the two most far-left political formations. The M5S occupies the centre of the left-right
continuum, with a position much closer to the PD’s than to other centre-right parties’ one.
The complexity of the strategic situation that emerged from the election of March
2018 was reflected by the length of the government bargaining process: 89 days from
the election; 69 from the resignation of the incumbent prime minister Gentiloni. The M5S
played a prominent position in such a process, negotiating both with centre-left and centre-
right parties in order to form a cabinet. Finally, after long negotiations, the two “winners”
of the election – the M5S and LN – reached an agreement to form a coalition government
led by Giuseppe Conte, a “neutral” prime minister with no previous political experience.9
The bargaining complexity of the formation opportunity which eventually led to the
Conte cabinet is also captured by our models, which assign very low choice probabilities
to the different potential governments that could have formed after the election. Figure 3
plots these probabilities for six plausible majority alternatives, all including the M5S. The
two models predict different outcomes: the first one is a coalition between the M5S and
the centre-right FI, the second one is a combination between the centre-left pre-electoral
alliance and the M5S. Despite their differences, both the models strongly underestimate
an agreement between the far-right LN and the centrist M5S, which in the end resulted to
be the government that actually formed.
9
Between 2013 and 2018, the M5S increased its vote share from about 25 per cent to more than
32 per cent. The LN moved from about 4 per cent to 17 per cent.
110.15
Predicted choice probability
0.10
0.05
0.00
D
I
U
cI
(Fo S.LN
S.F
d)
S.P
E
I.N
D.+
rme
M5
I.FD
M5
M5
S.P
N.F
M5
S.L
M5
Potential government
Model M1 M2
Note: The bars show the predicted probability of different potential governments following the
elections of March 2018 according to the estimates of the conditional logit models presented in the
paper. Alternatives are ordered according to the predictions of M1.
Figure 3: Predictive probabilities of different potential governments following the
election of March 2018
4 Party competition in the 2018 elections
4.1 The expert survey methodology and the left-right dimension
Following the research methodology developed by Laver and Hunt (1992) and Benoit and
Laver (2006), a survey among Italian experts was fielded in the aftermath of the Italian
general election of March 4, 2018. We asked political experts to locate the most significant
political parties – those obtaining at least one per cent of the popular vote (see Table 1)
– on nine substantive policy issues, as well as on the general left–right dimension by using
20-point scales. The dimensions in the survey measure parties’ support for public spending
vis-à-vis lower taxes (“Taxes v. Spending”), state regulation of the market (“Deregulation”),
liberal policies on matters such as abortion, gay rights and euthanasia (“Social”), integration
of immigrants (“Immigration”), environmental protection (“Environment”) and territorial
decentralization of decision-making (“Decentralization”). The survey also includes three
dimensions dealing with parties’ positions on specific aspects of European politics: the
scope of EU intervention (“EU authority”), its peacekeeping role (‘EU neutrality”) and the
12role of the European Parliament and national governments as democratic accountability
mechanisms (“EU accountability”). For each of these nine policy dimensions, experts were
also asked to locate each party on a scale measuring the importance or salience of the
dimension for that party. This scale ranges from “1” (not important at all) to “20” (very
important).10
The expert survey methodology is grounded in the spatial approach to party competi-
tion, which relies on the assumption that policy spaces can be used to describe preferences
and choices of relevant political actors. While many configurations of policy spaces are
possible – measuring actors’ preferences on specific dimensions, such as economic, social,
immigration policy or EU-relates issues (Bakker et al. 2015; Benoit and Laver 2006) – in
practice these positions tend to be summarised on a single “left–right” continuum that has
been widely used to measure the positions and movements of political parties in Europe
and to test spatial models of political competition in a comparative perspective (Gabel and
Huber 2000; Laver and Budge 1992). Despite the fact that a widely understood “left–right”
continuum has been used to measure parties’ and voters’ ideological positions in the West-
ern political world, a widespread disagreement exists on the conceptual foundations of this
dimension and the way to measure it (Benoit and Däubler 2017).
Conceptually, there are two basic approaches to define the left-right divide. The first
one specifies the substantive content of the left-right a priori based on theoretical reasoning.
Following this approach, for example, the Comparative Manifestos Project RILE (RIght-
LEft) scale pre-defines political categories as being left or right to code party manifestos
(Budge et al. 2001). In the second approach, “the left-right dimension is defined inductively
and empirically as the super issue that most constrains parties’ positions across a broad
range of policies” (Gabel and Huber 2000, p. 96). The content of the left-right continuum
can consequently only be inferred a posteriori in a given context.
Methodologically, several techniques – such as mass surveys, expert surveys, and con-
tent analysis of party manifestos – have been used to estimate party positions (Laver
2001). More recently, a new area of research has started to apply automated and sta-
tistical approaches to scale positions from political “text as data” (Grimmer and Stewart
2013). Among these techniques, expert surveys occupy a prominent position: they are a
relatively quick and costless way of collecting data on parties; they offer the researcher sev-
eral possibilities to assess confidence in the accuracy of estimates; finally, this methodology
provides scores for individual parties even when they contest the elections as members of
pre-electoral coalitions.11 For these characteristics, expert surveys have been often used to
validate left-right parties’ placements obtained through other methods.
In the expert survey methodology used in this paper, expert placements of parties
10
Experts were selected from members of the Italian Political Science Association (SISP). We
sent email invitations to 316 experts, 71 of whom completed the questionnaire, with a response
rate of about 22.5 per cent.
11
This is particularly important for the Italian case, because since 1994 parties compete in pre-
electoral alliances which often present a joint electoral platform.
13on the left-right dimension are derived without specifying in advance what this should
mean. Previous analyses of party placements on this dimension in a number of Western
and European countries have shown that ideological positions on the left-right continuum
can be predicted from policy locations on other more specific policy scales included in the
survey, and in particular from the economic left-right (“Taxes v. Spending”) and the social
conservatism dimension (“Social”) (Benoit and Laver 2006, pp. 132-136).
4.2 Assessing the structure of the Italian policy space, 2018
In order to assess the structure of the policy space in the aftermath of the Italian general
election of March 2018, we will first use the expert survey data introduced above to identify
which policy dimensions appear to be more important in Italy. Second, we will analyse
the patterns of correlation between party positions on different policy dimensions revealed
by the experts in order to identify the underlying axes of political competition.
As noted above, experts were asked to locate parties on a saliency scale, measuring
the importance of a particular dimension for that party. Party salience scores enable us to
understand which dimensions are the most relevant. We measure the overall importance
score for each policy dimension in the 2018 general election by computing, for each issue,
the mean of the party-specific salience scores and weighting it by the vote share received
by each party. Figure 4 reports the overall importance score of each dimension, as well as
its 95 per cent confidence intervals.
As Figure 4 shows, the “Immigration” dimension, which measures attitudes towards
the integration of immigrants was judged by our sample of experts to be the most im-
portant issue shaping political competition in the Italian election of March 2018. The
second and third most salient dimensions are related to two EU issues: parties’ propen-
sity to increase/reduce the set of areas subject to European intervention (“EU authority”)
and the relative powers of the European institutions vis-à-vis the national ones (“EU ac-
countability”). These dimensions received on average higher importance scores than the
two dimensions dealing with economic policy (“Taxes v. Spending” and “Deregulation”),
which are the fourth and the fifth most important issues. In comparison to past waves
of expert surveys fielded in Western European countries – which revealed that economic
issues and social policy were most often the top-rated dimensions (Benoit and Laver 2006)
– these data indicate a strong change in the relative weight given by Italian parties to new
cultural conflicts linked to the opening up of national borders (Kriesi et al. 2006, 2008).
These data also suggest a significant change in comparison to the spatial structure of party
competition characterizing Italy in 2013, whereas “EU authority” and “Taxes v. Spending”
were the two most important dimensions (Di Virgilio et al. 2015; Giannetti, Pedrazzani
and Pinto 2017).
Having identified the most salient policy domains in the period under study, we now
turn to providing a more synthetic interpretation of the underlying dimensional structure
14Immigration
EU Authority
EU Accountability
Taxes v. Spending
Dimension
Deregulation
EU Neutrality
Social
Decentralization
Environment
0 5 10 15 20
Salience
Note: Saliency score are weighted by parties’ vote shares. Error bars represent 95 per cent confi-
dence intervals.
Figure 4: Saliency scores by dimension
of Italian party competition in 2018. We assess the dimensionality of political competition
in Italy by using exploratory factor analysis (Benoit and Laver 2006). Factor analysis is
a statistical “data reduction” technique that allows us to describe the variability among
a large set of observed variables in terms of few unobserved underlying factors. Each of
the extracted factors can then be interpreted substantively by looking at those original
variables that “load” on the factor. The main results of this technique applied to the
experts’ placements of Italian parties’ policy positions are illustrated in Table 2, which
reports the factor loadings of each policy domain on the latent dimensions for a three-
factor configuration.12
As reported by Table 2, the first and most important factor emerging from the analysis
is associated with the dimensions dealing with immigration, EU authority, EU accountabil-
ity, and social conservatism. Thus, this latent factor captures parties’ attitudes towards old
(cultural liberalism) and new cultural issues, among which European integration and im-
migration result to be prominent. Such issues correspond to the new political and cultural
12
To obtain factors that are more easily interpretable, we used varimax rotation, which is one
of the most commonly employed rotation options. A scree test suggests that a three-factor con-
figuration best fits our data. Only factors with eigenvalues greater than unity are conventionally
interpreted.
15Table 2: Dimensional analysis of the Italian policy space, 2018
Factor1 Factor2 Factor3
Taxes v. Spending 0.224 0.714 0.125
Social 0.664 0.345 0.351
Deregulation 0.070 0.785 0.045
Environment 0.255 0.590 0.376
Decentralization -0.028 -0.091 -0.536
Immigration 0.737 0.308 0.194
EU Authority 0.839 0.066 -0.032
EU Accountability 0.616 0.346 -0.107
EU Neutrality 0.491 -0.092 -0.416
Eigenvalue 3.135 1.157 0.195
Proportion explained 0.48 0.36 0.16
Note: Exploratory factor analysis with a three-factor con-
figuration weighted by the vote share received by each party.
N = 639. Variable loadings higher than 0.5 are in bold.
axes of competition linked with globalization and the rise of the new populist radical right
parties (Kriesi et al. 2006, 2008). This factor can be therefore labelled as a “demarcation-
integration” dimension. The second factor emerging from the analysis is related almost
exclusively with economic factors, configuring therefore a traditional economic left-right
divide, with the addition of environmental protection related issues. Such configuration
of the policy space represents a significant change from the past (see Di Virgilio et al.
2015). The “demarcation-integration” dimension has gained importance in comparison to
2013 as it has been extensively used by new parties, such as the M5S and parties such as
the League that underwent a significant change under the leadership of Matteo Salvini, as
a way to mobilise their electorate. Meanwhile, we observe a decline in the importance of
the traditional economic left-right dimension, which does not represent any more the main
axis structuring party competition in Italy.
This configuration of the policy space has been estimated considering parties’ policy
positions assessed by Italian experts on nine specific dimensions. However, experts were
also asked to locate parties on a general left-right dimension, taking all aspects of party
policy into account. The left-right continuum – which has been used in the statistical anal-
ysis described above – is only moderately correlated with the two main axes of competition
identified by factor analysis (0.56 and 0.52 respectively). Moreover, a regression of expert
left-right scores from party positions on the nine specific issues – whose results are not
reported here for sake of brevity – reveals that the former can be predicted quite well ex-
clusively from parties’ placements on social, economic and immigration dimensions. While
in the mind of our experts parties’ positions on the left-right spectrum can be constructed
from their positions on economic, social and immigration policy, the pattern of correlations
16between party positions along different policy dimensions depicts a very different scenario
that hardly fits with the one represented by the left-right continuum.
4.3 A two-dimensional map of the Italian party system, 2018
Once identified the underlying structure of the policy space in the 2018 Italian election, as
a further step in our analysis we estimate parties’ positions on the two first dimensions of
competition emerging from factor analysis (i.e. those with eigenvalues higher than one).
Figure 5 presents a two-dimensional map of this policy space. The horizontal axis identifies
a “demarcation-integration” dimension (factor 1), while the vertical axis represents the
economic left-right (factor 2). We estimate each parties’ preferences using mean regression
scores from factor analysis. The figure also shows a division of the policy space into regions
occupied by each party – denoted by dashed lines. These portions of the space – known as
“Voronoi tessellations” (Benoit and Laver 2006) – define regions closer (in Euclidean terms)
to a given party than to any other party. The segments connecting the ideal preferences
of the three main parties involved in the negotiations for the formation of the Conte
government – the PD, M5S and LN –represent the Euclidean distance which separates the
three points.
Figure 5 highlights a tripolar configuration of the basic structure of the space, in which
the PD, the M5S and the LN are located in three different quadrants of the graph. The
PD supports an integration strategy and it is on the centre-left of the economic left-
right dimension. Conversely, both the M5S and the LN promote a demarcation strategy.
However, they are placed on two different sides of the economic left-right continuum: the
M5S on the left and the LN on the right. Contrary to what happens in other countries (see
Kriesi et al. 2006, 2008), we do not observe a tripolar structure in which the “populist”
right constitutes a new third pole, but a configuration in which there are two “populist”
poles – one on the right and another one on the left – opposed by a mainstream centre-
left party. The figure also shows that the positions of the three parties configuring the
poles structuring the Italian policy space in 2018 vary more strongly with respect to the
“demarcation-integration” dimension than with respect to the economic left-right. The
latent dimension linked to old and new cultural aspects is therefore not only the most
salient, but also the most polarising one.
The policy space emerging from the election of March 2018 profoundly differs from
that registered in 2013 (see Di Virgilio et al. 2015). FI, i.e. the centre-right party founded
and still guided by Silvio Berlusconi, ceased to be a strategic relevant actor in the Italian
political competition. Moreover, EU related issues have been incorporated in a cultural
dimension together with immigration and social conservatism and, finally, the economic
left-right is no more the main dimension structuring party competition. These elements
represent evidence of a structural reshaping of the Italian party system, a phenomenon
registered in other Southern European countries following the Great Recession (Bosco and
172
1
FI
+EU LN
NcI
Economic left-right
2.000
0
FdI
1.099
PD
1.549 M5S
LeU
-1 PaP
-2
-2 -1 0 1 2
Integration-Demarcation
Vote share a 10 a 20
a 30
Note: Parties’ positions are mean regression scores from factor analysis. Label size is proportional
to vote share. Numbers indicate the Euclidean distance between the three parties connected by
the segments. Dashed lines denote Voronoi tessellations.
Figure 5: A two-dimensional map of the Italian policy space, 2018
Verney 2012, 2016).
Finally, the policy space represented in Figure 5 helps to understand the outcome of the
government formation process which our statistical models fail to predict. The length of the
segments connecting the three main parties involved in the Conte government negotiations
clearly supports a coalition between M5S and LN compared to a different one formed by
M5S and PD. The M5S-LN combination constitutes, on the one hand, a minimal winning
coalition in both the branches on the Italian parliament; on the other hand, it minimises
the ideological distance (in Euclidean terms) between the parties in the coalition. The
coalition between M5S and LN results therefore the most “rational” combination according
to both office-seeking and policy-seeking incentives.
185 Conclusion
Despite the fact that the ideological distance between parties along the left-right dimension
represents a significant driver of government formation in Italy between 2001 and 2018, it
fails to predict the government that formed after the Italian general election of March 2018.
In this paper we attempted to show that this deviation between theoretical predictions and
real outcomes is mainly a consequence of the inappropriateness of the general left-right
continuum in describing political competition in Italy and it is not due to a failure of the
spatial approach to coalition formation when applied to the Italian case.
Using data from an expert survey fielded by the authors in the aftermath of the election
of March 2018, we found that political competition in Italy can be better explained by a
two-dimensional policy space. The first dimension is identified by immigration, EU related
issues and social conservatism, configuring therefore a “demarcation-integration” contin-
uum. The second dimension measures the classic economic left-right. On the basis of such
two-dimensional approach, we were able to propose an explanation of the Conte govern-
ment formation. Such outcome appeared puzzling to many commentators who judged the
M5S and the NL as quite distant along the left-right dimension. Indeed, the two parties
supporting the Conte government are those that minimise the Euclidean distance in the
two-dimensional spatial configuration emerging through our analysis of the policy space
based on expert survey data.
A number of important implications that may contribute to future research can be
drawn from our results. First, our study reveals that the substantive policy dimensions
that experts have in mind when defining left and right are mainly associated to socio-
economic factors. However, there is evidence that in most countries, and in particular in
Southern European ones after the Great Recession, the economic dimension is declining in
importance, while new cultural divisions are becoming in general more salient. Researchers
should therefore be aware of this discrepancy when designing cross-national studies which
employ parties’ positions on a general left-right scale. Second, highlighting the emergence
of a new cultural dimension and the related success of two powerful anti-establishment
parties, our research may contribute to those studies aiming to evaluate the impact of the
economic crisis and EU policy on democratic representation in European countries. The
obvious limitations of our study are inherently related to a framework that mainly rely
on policy factors as opposed to valence issues to analyse party competition. According
to many scholars, valence issues seem to have acquired more and more importance in
shaping political competition. Notwithstanding these limitations, we are confident that
our work demonstrates that the spatial approach is still a powerful tool to study coalition
governments’ formation.
19References
Anelli, Massimo, Italo Colantone, Massimo Pulejo, and Piero Stanig. 2018. “Italy just
voted for two very different kinds of populism.” The Washington Post, March 28.
Retrieved from: goo.gl/ntrSPW.
Austen-Smith, David and Jeffrey Banks. 1988. “Elections, coalitions, and legislative out-
comes.” American Political Science Review 82(2): 405–22.
Axelrod, Robert M. 1970. Conflict of interest: A theory of divergent goals with applications
to politics. Chicago: Markham Publishing Company.
Bäck, Hanna, and Patrick Dumont. 2007. “Combining Large-n and Small-n Strategies:
The Way Forward in Coalition Research.” West European Politics 30(3): 467–501.
Bakker, Ryan et al. 2015. “Measuring Party Positions in Europe: The Chapel Hill Expert
Survey Trend File, 1999-2010.” Party Politics 21(1): 143–52.
Baldini, Gianfranco. 2011. “The Different Trajectories of Italian Electoral Reforms.” West
European Politics 34(3):644–63.
Benoit, Kenneth, and Michael Laver. 2006. Party Policy in Modern Democracies. London,
New York: Routledge.
Benoit, Kenneth, and Thomas Däubler. 2017. “Estimating Better Left-Right Positions
Through Statistical Scaling of Manual Content Analysis.” Working paper.
Black, Duncan. 1958. The Theory of Committee and Elections. Cambridge: Cambridge
University Press.
Bosco, Anna, and Susannah Verney. 2012. “Electoral Epidemic: The Political Cost of
Economic Crisis in Southern Europe, 2010-11.” South European Society and Politics
17(2): 129–54.
Bosco, Anna, and Susannah Verney. 2016. “From Electoral Epidemic to Government
Epidemic: The Next Level of the Crisis in Southern Europe.” South European Society
and Politics 21(4): 383–406.
Budge, Ian, Hans-Dieter Klingemann, Andrea Volkens, Judith Bara, and Eric Tanenbaum.
2001. Mapping Policy Preferences: Estimates for Parties, Electors, and Govern-
ments 1945–1998. Oxford: Oxford University Press.
Chiaramonte, Alessandro, and Roberto D’Alimonte. 2018. “The New Italian Electoral
System and Its Effects on Strategic Coordination and Disproportionality.” Italian
Political Science 13(1): 8–18.
20Chiaramonte, Alessandro, and Vincenzo Emanuele. 2017. “Party System Volatility, Regen-
eration and de-Institutionalization in Western Europe (1945-2015).” Party Politics
23(4): 376–88.
Chiaramonte, Alessandro. 2015. “The Unfinished Story of Electoral Reforms in Italy.”
Contemporary Italian Politics 7(1): 10–26.
Cotta, Maurizio, and Luca Verzichelli. 2007. Political Institutions in Italy. Oxford, New
York: Oxford University Press.
Curini, Luigi, and Luca Pinto. 2013. “Government Formation under the Shadow of a Core
Party: The Case of the First Italian Republic.” Party Politics 19(3): 502–22.
Curini, Luigi, and Luca Pinto. 2017. L’arte Di Fare (e Disfare) I Governi: Da De Gasperi
a Renzi, 70 Anni Di Politica Italiana. Milano: Egea.
De Swaan, Abram. 1973. Coalition theories and cabinet formations: A study of formal
theories of coalition formation applied to nine European parliaments after 1918. Am-
sterdam: Elsevier.
Debus, Marc. 2009. “Pre-Electoral Commitments and Government Formation.” Public
Choice 138(1–2): 45–64.
Di Virgilio, Aldo, Daniela Giannetti, Andrea Pedrazzani, and Luca Pinto. 2015. “Party
Competition in the 2013 Italian Elections: Evidence from an Expert Survey.” Gov-
ernment and Opposition 50(1): 65–89.
Downs, Anthony. 1957. An Economic Theory of Democracy. New York: Harper.
Gabel, Matthew J., and John D. Huber. 2000. “Putting Parties in Their Place: Infer-
ring Party Left-Right Ideological Positions from Party Manifestos Data.” American
Journal of Political Science 44(1): 94–103.
Giannetti, Daniela, Andrea Pedrazzani, and Luca Pinto. 2017. “Party System Change in
Italy: Politicising the EU and the Rise of Eccentric Parties.” South European Society
and Politics 22(1): 21–42.
Golder, Sona N. 2006. The Logic of Pre-Electoral Coalition Formation. Columbus: Ohio
State University Press.
Grimmer, Justin, and Brandon M. Stewart. 2013. “Text as Data: The Promise and Pitfalls
of Automatic Content Analysis Methods for Political Texts.” Political Analysis 21(3):
267–97.
Kriesi, Hanspeter et al. 2006. “Globalization and the Transformation of the National
Political Space: Six European Countries Compared.” European Journal of Political
Research 45(6): 921–56.
21You can also read
