Introduction
Over the years, research on coalition politics and foreign policy has underlined the crucial role of junior partners, that is, those parties holding a smaller share of seats in parliament and cabinet posts (Kaarbo, Reference Kaarbo1996; Ozkececi-Taner, Reference Ozkececi-Taner2005; Clare, Reference Clare2010; Oppermann and Brummer, Reference Oppermann and Brummer2014). However, junior partners do not often manage to steer coalition foreign policy towards their preferred direction. At this point, they are confronted with a burning decision. On the one hand, they can choose to keep on being part of the coalition cabinet and, perhaps, try to influence the specific foreign policy from within. On the other hand, they may decide to quit, abandoning their positions in power: a move that might satisfy their own voters but also prevent these parties from affecting the policymaking process. The latter decision inevitably produces the termination of the government. On the basis of empirical evidence (Kaarbo, Reference Kaarbo2012), losing a coalition conflict on foreign policy does not seem to be a sufficient reason for junior partners to defect in most cases. Regrettably, the causal mechanisms leading to this outcome are unexplored in the literature on coalition politics and foreign policy.
The article aims to fill this literature gap, investigating what motivates junior partners to remain in the government after losing a coalition conflict on foreign policy. In other words, I address the following question: why do small parties stay in the cabinet in spite of a disliked foreign policy decision? In order to explain this outcome, I analyse and compare two cases of Italy's participation in Military Operations Abroad (MOA): Operation Allied Force (OAF) in Kosovo (1999) and Operation Unified Protector (OUP) in Libya (2011).
Notably, the increasing involvement in MOAs represents the most transformative element in post-Cold War Italian foreign policy (Coticchia, Reference Coticchia2011; Ignazi et al., Reference Ignazi, Giacomello and Coticchia2012; Carati and Locatelli, Reference Carati and Locatelli2017). A few works underline how such a development was fostered by a bipartisan consensus between centre-left and centre-right parties (Calossi and Coticchia, Reference Calossi and Coticchia2009; Coticchia and Vignoli, Reference Coticchia and Vignoli2018; Vignoli, Reference Vignoli2019). However, during the so-called ‘Second Republic’, centrist parties were invariably in government with extreme (left and right) parties, holding a more sceptical position concerning MOAs. This created tensions between coalition partners that never provoked the collapse of a government, nonetheless. Among the vast number of MOAs in which Italy took part, I analyse OAF and OUP because of their similarities in terms of scopes, operational features and institutional framework. Furthermore, junior partners' extent of contestation has been particularly high in both cases. An element of variation consists in the ideological make-up of the coalitions and, consequently, of the extreme junior partners in power: the communist Partito dei Comunisti Italiani (PdCI) and green Federazione dei Verdi (FdV) in a centre-left coalition in the case of Kosovo, and the far-right and autonomist Lega Nord (LN) in a centre-right coalition in the case of Libya.
In line with Coticchia and Davidson (Reference Coticchia and Davidson2018), I hypothesize two causal mechanisms leading to this outcome: lack of salience attributed by the junior partner to security policy and fear of being punished by voters for the collapse of the government. On the basis of extensive qualitative empirical material, including interviews to policymakers, newspaper articles, memoirs and reports of the parliamentary debates, I highlight how these two mechanisms played out differently in the two cases. On the one hand, extreme-left and pacifist parties considered the opposition to the intervention in the Kosovo war as an extremely salient issue but were reluctant to be labelled as irresponsible coalition partners, paving the way for the return of the centre-right coalition in power. On the other hand, LN did not consider participation in the air strikes against the Libyan regime as a crucial issue per se and, conversely, was not afraid of making the government fall for other reasons. In line with Coticchia and Vignoli (Reference Coticchia and Vignoli2018), this finding suggests a variation in extreme parties' opposition to troop deployments during this period in Italy: while extreme-left parties rooted their dissent on profound pacifist beliefs, LNs had a more instrumental approach.
The article is structured as follows. First, I review the literature on coalition politics and foreign policy, focusing on studies underlining the relevance of junior partners in the decision-making process. I subsequently outline the two aforementioned mechanisms leading to junior partner's decision to stay in government in spite of a disliked foreign policy outcome. I then proceed explaining the criteria employed for the case selection. In the following section, I describe the qualitative material collected for the paper. Through these data, I examine the two cases, finding evidences of the mechanisms. Finally, in the conclusion, I resume and discuss the findings and their implications.
Coalition cabinets, junior partners and foreign policy
Within the broad academic debate on the democratic peace theory, controversy about the effect of government's composition in parliamentary democracies on states' propensity to be involved in conflict exists. On the one hand, according to a structural interpretation of the democratic peace theory (Maoz and Russett, Reference Maoz and Russett1993) and veto player theory (Tsebelis, Reference Tsebelis2002), coalition cabinets are supposed to be more dovish than single-party governments as they are more constrained by other institutional actors. On the other hand, following the diffusion of responsibility approach (Powell and Whitten, Reference Powell and Whitten1993), coalition cabinets are presumed to be more likely to involve their own countries into a military conflict than single-party governments as they can distribute domestic audience costs across parties. Most of the quantitative tests of these theories do not find any statistically significant distinction between single-party executives and coalition cabinets (Ireland and Gartner, Reference Ireland and Gartner2001; Reiter and Tillman, Reference Reiter and Tillman2002). Prins and Sprecher (Reference Prins and Sprecher1999) instead discover that coalition governments are more likely to reciprocate military disputes. To the contrary, complementary qualitative studies suggest that coalition cabinets are more dovish than single-party governments (Auerswald, Reference Auerswald1999; Elman, Reference Elman2000).
Subsequently, studies on coalition politics and foreign policy moved in two directions: exploring a wider range of foreign policy events other than war and distinguishing coalitions by number of parties, ideological fractionalization and parliamentary strength. Kaarbo and Beasley (Reference Kaarbo and Beasley2008) discover that coalition cabinets are neither more peaceful, nor more aggressive than single-party cabinets but more extreme in their foreign policy behaviour. Clare (Reference Clare2010) find that, while cohesive coalitions are as hawkish as single-party governments, ideologically fractionalized coalitions present a different propensity to initiate a military dispute according to the position of the outlier party on the left-right axis. Beasley and Kaarbo (Reference Beasley and Kaarbo2014) highlight that the number and relevance of parties, parliamentary strength and the presence of critical small parties are crucial variables to explain coalitions' foreign policy behaviour in terms of commitment, engagement and aggressiveness.
These more recent works emphasize the relevance of junior partners. For instance, Clare (Reference Clare2010: 966) argues that junior partners ‘can exercise powerful, even disproportionate, influence on foreign policy’. However, scholars considerably disagree on which factors increase junior partners' impact on coalition foreign policymaking. Kaarbo (Reference Kaarbo1996) identifies the arena in which decisions are taken, intra-party cohesion and strategy as crucial variables to explain junior partners' influence on the final decision. Ozkececi-Taner (Reference Ozkececi-Taner2005) emphasizes that junior partners have to perceive an issue as salient and develop a distinct position in order to play a constraining role in the government in foreign policy. Oppermann and Brummer (Reference Oppermann and Brummer2014) highlight that control of key ministries and departmental autonomy within the cabinet decisively shape the way in which junior partners are able to influence foreign policymaking. Brommesson and Ekengren (Reference Brommesson and Ekengren2019) point out how the margin of manoeuvre conceded by the senior partner has a decisive explanatory power in increasing or decreasing junior partner's influence.
However, junior partners' influence should not be overestimated. In fact, they often end up being the loser in the conflict within the coalition. In his seminal volume on foreign policymaking in coalition cabinets, Kaarbo (Reference Kaarbo2012) finds that in most cases junior partners' attempts to shape foreign policy outcomes were unsuccessful. Some cases of lack of influence are surprising considering the above-mentioned conditions. For instance, in the Netherlands, the small but pivotal Democrats 66 was able to delay but not prevent the government's much-contested decision to deploy troops in an extremely dangerous area of Afghanistan in 2006. In this sense, focusing on Swedish foreign policy, Brommesson and Ekengren (Reference Brommesson and Ekengren2019: 14) conclude that junior partner's influence was present ‘but only to a limited extent’ and related to symbolic issues rather than real policy outcomes.
Facing a disliked foreign policy decision, a junior partners is confronted with a tough choice. On the one hand, it may decide to keep on staying in the cabinet. Such a decision can be read as a defeat but allows the junior partner to influence the implementation of the policy at a later stage. Furthermore, it might exchange its loyalty on foreign policy with for a more favourable agreement on a policy which it attaches more importance to. On the other hand, the junior partner may defect, contributing to the termination of the cabinet. Pursuing this route, it shows ideological consistency to voters but also loses its position in power and, consequently, the ability to shape policies. Existing literature seems to suggest that the former strategy is more likely than the latter. In other words, a coalition conflict on foreign policy seldom provokes the collapse of the government. For example, out of 12 cases of conflict included in Kaarbo (Reference Kaarbo2012), only one directly caused the departure of the junior partner and the collapse of the entire cabinet. In 1995, the Turkish Republican People's Party decided to abandon the coalition government in which it was the junior partner as a consequence of the signing of a custom union treaty with the EU.
The literature on coalition politics and foreign policy has paid limited attention to reasons pushing a junior coalition partner to remain in the government despite an undesired foreign policy outcome. This puzzle deserves further attention as the willingness not to defect may significantly decrease junior partner's hijacking power in influencing the final outcome. Kaarbo (Reference Kaarbo2012) loosely refers to ‘political calculations’ linked to office-seeking and vote-seeking dynamics. Analysing the case of pacifist extreme-left parties and military operations in Italy, Coticchia and Davidson (Reference Coticchia and Davidson2018) underline the crucial role of salience attributed to the issue and fear of being blamed by voters. The article extends this work to a new case, contestation regarding the participation in the multilateral intervention in Libya in 2011, to provide a stronger test of these two factors.
Hypotheses
In line with Coticchia and Davidson (Reference Coticchia and Davidson2018), I hypothesize two elements underlying junior partner's decision to stay in government despite a disliked foreign policy outcome, and in particular, the decision to participate in a military intervention: low salience attributed to the issue and fear of being blamed by voters for the collapse of the cabinet. These two explanations are neither exclusive nor exhaustive: they may verify in conjunction and be complemented by other factors. However, they provide useful insights in explaining the phenomenon under investigation.
Salience refers to the importance attributed by a political actor such as parties, individual legislators or even voters to a specific policy issue. In particular, parties can consider a policy issue as important, responding to factors such as voters' preferences and position in the party system (Ezrow et al., Reference Ezrow, De Vries, Steenbergen and Edwards2011; Klüver and Spoon, Reference Klüver and Spoon2016). Various studies on coalition politics and foreign policy hypothesize salience to have a decisive impact on junior partner's influence in the decision-making process and the final policy outcome (Kaarbo, Reference Kaarbo1996; Ozkececi-Taner, Reference Ozkececi-Taner2005; Coticchia and Davidson, Reference Coticchia and Davidson2018). This may happen through the obtainment of the ministries of Foreign Affairs and Defence, that is in turn a demonstration of the salience attributed to this policy area (Bäck et al., Reference Bäck, Debus and Dumont2011; Oppermann and Brummer, Reference Oppermann and Brummer2014).
In these works, salience is conceptualized in relative rather than absolute terms. In other words, junior and senior partners are compared in terms of how much importance they attribute to a specific foreign policy issue. Kaarbo (Reference Kaarbo1996: 513) argues that ‘how salient is an issue to both the junior and senior party policy objectives may affect junior party influence’. If the junior partner considers a certain foreign policy issue as more salient than the senior partner does, it is expected to exert more influence in the decision-making process. The mechanism underlying this hypothesis is that, in such a situation, the junior partner engages in stronger and more consistent efforts to affect the decision-making process (Kaarbo, Reference Kaarbo1996). To put it bluntly, higher salience leads to more aggressive strategies of influence, including the threat to leave the cabinet.
Perceiving that the issue is deemed as important by its own voters, the junior partner is motivated to impose its own position on foreign policy to the whole cabinet in order not to lose votes in the following elections. As Brommesson and Ekengren (Reference Brommesson and Ekengren2019: 5) claim, ‘successful efforts to gain influence are probably accompanied by a strategic or ideological judgment by the junior party that the government/own party can gain support among voters’. To the contrary, in cases in which the salience is lower, a junior partner is not presumed to be as committed to influence the foreign policy outcome. This depends on the fact that its voters also do not consider the issue as salient, and consequently, will vote again the party disregarding of the government's final decision.
The salience hypothesis has ramifications for the junior partner's decision to remain in government in spite of a disliked foreign policy outcome too. Low salience attributed to a foreign policy issue is presumed to increase extreme junior partner's likelihood to stay in power. According to Brommesson and Ekengren (Reference Brommesson and Ekengren2019: 5), ‘if the (junior) party does not find the policy to be of great importance they are more likely to stay loyal to the government’. In addition, Coticchia and Davidson (Reference Coticchia and Davidson2018: 153) claim that the ‘radical party might be willing to pay this price for issues of the utmost importance but is unlikely to do so for issue of lower salience’.
Various studies demonstrate how in Italy security and defence policy has been consistently marginalized in the political and public debate (Coticchia, Reference Coticchia2011; Ignazi et al., Reference Ignazi, Giacomello and Coticchia2012). Therefore, military operations are not supposed being attributed high salience by Italian parties. This decreases the incentives for junior coalition partners to leave the cabinet due to disagreements on the cabinet's decisions on this issue.
H1: Low salience attributed to a foreign policy issue increases junior partner's propensity to stay in government.
It has just been assumed that voters reward junior partners on the basis of the decisions taken by the cabinet on salient issues. However, the electorate may highly value the stability of the government per se. More durable governments are thought to have more time to implement their policies and, consequently, being more effective. Therefore, the median voter may punish a junior partner for making the government collapse. In a bipolar and highly ideologically polarized party system, in which left and right coalition governments alternate, extreme junior partners' voters may prefer stability over a foreign policy outcome in line with their views as well. In fact, the collapse of a cabinet may lead to the rise to power of another cabinet that is very distant from their ideological preferences. As Coticchia and Davidson argue (Reference Coticchia and Davidson2018: 153), ‘In a bipolar system, we expect median voters to react to centre-left collapse by moving their vote to the centre-right. We can also expect even dedicated left-leaning voters to punish parties that force government collapse by moving their vote to another party on the left (not complicit in the collapse)’. Therefore, in such a party system, extreme junior partners are expected to be concerned about taking the blame for the collapse of the cabinet. This mechanism acts as an incentive for not leaving the cabinet due to a disagreement over foreign policy.
In Italy, during the so-called ‘Second Republic’, centre-left and centre-right cabinets perfectly alternated (Verzichelli and Cotta, Reference Verzichelli, Cotta, Müller and Strøm2000). Relatively large centre-left and centre-right parties had to coalesce with smaller parties, including extreme ones, to form cabinets.. Therefore, for instance, the defection of an extreme-left party could determine the collapse of a centre-left cabinet and the rise to power of a centre-right cabinet. For this reason, as said, junior extreme partners are expected to be not as prone to leave the government due to a disliked foreign policy outcome.
H2: Fear of being blamed as irresponsible by the electorate increases junior partner's propensity to stay in government.
The two mechanisms are presumed to manifest in various ways. Scarce interest in the specific foreign policy or inconsistent efforts to steer foreign policy towards a different direction are signs of lack of salience attributed to the issue. Furthermore, the manner in which a junior partner frames its contestation may be revealing about the extent salience it attaches to military interventions. Instrumental criticism concerning side effects may also be revealing of a lower extent of salience. The blame hypothesis regarding the perception of being labelled as unreliable government partners by voters is supposed to emerge by statements justifying the decision not to defect on the basis of such rationale. It may also emerge from the lack of willingness to leave the cabinet, disregarding the issue of conflict with the senior partner.
Case selection
In order to identify the reasons why junior partners do not defect in spite of the implementation of an undesired foreign policy, I examine two cases of disagreement between coalition partners in Italy on participation in MOA. The deployment of troops in missions outside national borders arguably constitutes the most remarkable development in Italian post-Cold War foreign policy (Coticchia, Reference Coticchia2011; Ignazi et al., Reference Ignazi, Giacomello and Coticchia2012). After the fall of the Berlin wall, Italy has taken part in an increasing number of MOAs around the world, turning ‘from a security consumer to a security provider’ (Walston, Reference Walston2007). Its armed forces have been present in all the major areas of crisis, including Kosovo, Afghanistan and Iraq.
Several studies explore this transformation in Italian security and defence policy, employing different theoretical approaches and analytical perspectives (Cladi and Webber, Reference Cladi and Webber2011; Rosa Reference Rosa2014). Scholars identify the quest for prestige as the main driver of the Italian contribution to multilateral operations (Carati and Locatelli, Reference Carati and Locatelli2017; Cladi and Locatelli, Reference Cladi and Locatelli2019). A few works investigate how political parties positioned on the issue, highlighting a substantial bipartisan consensus between the main centre-left (PDS-DS-PD) and centre-right parties (Forza Italia) (Calossi and Coticchia, Reference Calossi and Coticchia2009; Coticchia, Reference Coticchia2011; Ignazi et al., Reference Ignazi, Giacomello and Coticchia2012; Coticchia and Vignoli, Reference Coticchia and Vignoli2018; Vignoli, Reference Vignoli2019). Extreme parties (on the left and the right) were not instead part of this consensus and often expressed scepticism with regards to Italy's participation in MOAs. A strong pacifist culture and criticism regarding international institutions such as the UN and NATO, seen as too subjected to the United States' will, drive socialist and communist parties' contestation of peace and security operations (Calossi and Coticchia, Reference Calossi and Coticchia2009; Calossi et al., Reference Calossi, Calugi and Coticchia2013). At the extreme-right side of the spectrum, while the post-fascist Alleanza Nazionale saw in the military contribution to MOAs a source of national pride, the far-right, populist and autonomist LN criticized it also on the basis of an anti-American feeling at first and, later on, as a waste of taxpayers' money (Tarchi, Reference Tarchi and Schori Liang2007; Calossi and Coticchia, Reference Calossi and Coticchia2009; Verbeek and Zaslove, Reference Verbeek and Zaslove2015).
These inevitable divergences between centrist and extreme parties never directly led to the collapse of either a centre-left or centre-right cabinet. This is surprising considering that Italy is often quoted as a case of government instability. In fact, during its history as a republic, it has scored the highest rate of government change in Western Europe. The average cabinet duration is also well below the average (Curini and Pinto, Reference Curini and Pinto2017). Furthermore, conflicts between coalition partners have been a prominent source of government termination (Damgaard, Reference Damgaard, Strøm, Muller and Bergman2008).
In order to do explain this puzzling outcome, among the various MOAs in which successive Italian governments decided to participate, I analyse two cases: OAF (Kosovo, 1999) and OUP in Libya (2011).Footnote 1 The reasons are threefold.
First, the two military interventions present remarkable similarities in terms of scopes, chain of command and means. Both OAF and OUP had as their official goal the protection of civilians from the acts of violence perpetrated by two authoritarian regimes. In Kosovo, in the context of a civil war for the independence of the region, Serbian armed forces were responsible for thousands of killings among the Albanian ethnic minority. In Libya, militaries brutally repressed public protest against the regime and successively engaged in an armed conflict against rebels. Therefore, the so-called doctrine of ‘Responsibility to Protect’ (R2P) was employed to justify the use of force in these two interventions. The debate over the legitimacy of OAF paved the way for the development of this doctrine (Bellamy, Reference Bellamy2009). OUP is instead widely considered as a unique example of the application of R2P in practice (Patrick, Reference Patrick2011). Moreover, both interventions were conducted under the coordination of NATO.Footnote 2 However, it has to be pointed out that while OUP was authorized by the United Nations, through Security Council's Resolutions 1970 and 1973, OAF did not receive such legitimization. Finally, both these missions consisted of air strikes. In a little more than 3 months, over 10,000 strike sorties were done for OAF, while in 7 months, 9700 were done for OUP. Italy actively participated in the bombings in both cases (Coticchia, Reference Coticchia2011; Ignazi et al., Reference Ignazi, Giacomello and Coticchia2012). The national contribution to OAF was second only to the American one (Coticchia, Reference Coticchia2011). In total, 50 Italian warplanes were involved in the air strikes in Kosovo, for 1300 sorties (Coticchia, Reference Coticchia2011). Italy's contribution to OUP was significant as well. Italian air forces flied on the Libyan skies for up to 7300 h in total, conducting almost 500 missions. Only the United States, United Kingdom and France were more involved in the NATO campaign in Libya (Coticchia, Reference Coticchia2011).
Secondly, as suggested, junior coalition partners with completely different ideological leanings contested these two interventions. In the case of OAF, the communist PdCI and the Green FdV were strongly against the military intervention and, consequently, Italy's contribution to it. Both these parties shared a pacifist approach to foreign and security policy. PdCI rooted its scepticism towards the intervention in its anti-American attitude, and, conversely, a close relationship with the Serbian leadership. In the case of OUP, the far right and autonomist LN heavily criticized the decision to meddle in the Libyan civil war. Such criticism was mainly based on the fear that the collapse of the Gaddafi regime would have led to an increase in the influx of illegal migrants to Italy. These extreme parties are very peculiar but, to some extent, the sources of their opposition to these military interventions are shared by other like-minded parties in Western Europe. In fact, while anti-Americanism is a widespread feeling among many radical-left and communist parties, control of immigration is regarded as a fundamental issue by far-right parties (Fabbrini, Reference Fabbrini2002; Ignazi, Reference Ignazi2003)
These parties were (extreme) junior partners in the two coalitions that alternated in government in Italy during this period. Participation in OAF was in fact promoted by a centre-left cabinet led by Massimo d'Alema, a prominent figure in the senior partner Democratici di Sinistra, and composed of up to eight parties in total, bringing together former Christian democrats and communists. Contribution to OUP was instead decided by a centre-right coalition government, led by Silvio Berlusconi, and composed only of two parties: the conservative Popolo della Libertà (PdL) and LN indeed.Footnote 3 Notably, in both cases, junior partners were pivotal for the survival of the cabinet. Without the combined support of both PdCI and FdV, D'Alema centre-left cabinet would have lost its parliamentary majority at the Chamber of Deputies. LN could exercise an equal hijacking power on Berlusconi's centre-right cabinet. It is also worth noting that none of these junior partners held key cabinet portfolios, namely foreign affairs and defence, in both cases. During the Kosovo crisis, the liberal Lamberto Dini and the Christian democrat Carlo Scognamiglio were the Minister of foreign affairs and defence, respectively. During the Libyan crisis, both the Ministers of foreign affairs and defence, Franco Frattini and Ignazio La Russa, respectively, were affiliated to the senior partner PdL.
Thirdly, the cases present similar developments in the conflict within Italian coalition governments and an identical outcome for the survival of the two cabinets. In fact, junior partners did not eventually defect, making the government survive. After the end of OAF in Kosovo in June 1999, the D'Alema cabinet stayed in power for a few more months and, eventually, had to resign because of the defection of two small centrist parties, Cristiani Democratici Uniti and Unione Democratica per la Repubblica.Footnote 4 After the end of OUP, in November 2011, the fourth centre-right Berlusconi's coalition government collapsed in the face of a dramatic debt crisis. To sum up, the coalition conflicts over participation in these multilateral interventions did not provoke the termination of the government, as extreme parties decided to remain in the cabinet.
Therefore, this cas selection is based on the criteria of most similar cases in terms of the dependent variable (Seawright and Gerring, Reference Seawright and Gerring2008). In fact, as explained, the two cases present a similar outcome in the sense that the junior partner decided not to leave the government. However, at the same time, the two cases differ completely in terms of the junior partner's ideological leaning. This element allows to see how our hypothesized factors had a varying impact on the decision to remain in government, according to the ideological leaning of the junior partner involved.
I instead exclude from the analysis another prominent case of disagreement between coalition partners over security and defence policy in Italy during the ‘Second Republic’. After losing a vote at the Senate on the re-funding of military operation ISAF in Afghanistan and the expansion of a US military base in Vicenza, on 21 February 2007, the then Prime Minister Romano Prodi resigned. The resignation was then rejected by the President of the Republic Giorgio Napolitano. While the defeat was due to the defection of a few rebels, it is true that extreme-left parties were against the mission in Afghanistan (Coticchia and Davidson, Reference Coticchia and Davidson2018). However, this case is not included due to comparability issues. In fact, as said, the debate was not exclusively centred on contribution to an MOA but also pivoted around another security issue that was particularly controversial in the political and public debate. Moreover, ISAF presents a series of differences with respect to OAF and OUP. In fact, ISAF lacked such an explicit R2P justification. Furthermore, it was conducted on the ground, in a country not proximate to Italy.
Data
In order to investigate the dynamics leading to junior partners' permanence at the government, I rely on a series of qualitative data: elite interviews with relevant policymakers, memoirs of cabinet members, parliamentary debates and reports, and newspaper articles.
As the only original source of data, interviews deserve particular attention.Footnote 5 In total, I conducted 13 interviews: seven for the case of Kosovo and six for the case of Libya. I interviewed representatives belonging to both senior partners and junior partners. For example, in the case of Kosovo, I interviewed three representatives for PdCI, further three for the greens and one for Democratici di Sinistra. Instead in the case of Libya, I interviewed three representatives for LN and further three for Forza Italia. Interviews with junior partners' representatives are particularly useful to uncover their beliefs and perceptions. In fact, to some extent, both hypotheses are linked to perceptions about the salience of the issue and the potential electoral costs deriving from leaving the government. Interviews with senior partners' policymakers are employed to triangulate and assess if they shared the same impressions.
Newspaper articles, parliamentary debates and memoirs of relevant policymakers complemented the interviews, providing further empirical material for the reconstruction of the events and the analysis. Newspaper articles containing direct statements from junior partners' exponents are used to validate and strengthen the data contained in the interviews. Parliamentary debates and reports serve instead to assess the efforts made by a junior partner to influence the executives' position and their extent of commitment to monitor the development of the mission. Newspaper articles were retrieved from the digital archive of La Repubblica, the second most read Italian daily newspaper. Parliamentary debates were retrieved from the websites of the Chamber of Deputies and the Senate.
Empirical findings
Kosovo
Participation in MOAs proved a contentious issue within the centre-left coalition even before the beginning of the Kosovo crisis. In fact, on 9 April 1997, the communist party Partito della Rifondazione Comunista (PRC), providing essential parliamentary support for the centre-left cabinet, led at the time by Romano Prodi, voted against Operation Alba, aimed at restoring security and holding free elections in the near Albania. The mission was approved only thanks to the support of centre-right opposition parties, with the exception of LN.Footnote 6 However, a couple of days later, PRC returned to the ranks, guaranteeing its votes for the cabinet in a confidence vote. This event is relevant for the explanation of why extreme-left parties did not leave the cabinet one year later during the Kosovo crisis for two reasons. First, it shows the high extent of salience attributed by communists to the issue of participation in MOAs. Second, it highlights the emergence of a bipartisan consensus between the main centre-left and centre-right parties, forcing extreme-left parties not to defect in order to be able to steer the government's position towards a more dovish direction.
The split within PRC, provoking the collapse of the Prodi cabinet and the formation of a new centre-left cabinet, is also crucial to understand why junior coalition partners stayed in the government despite the decision to contribute to the intervention in Kosovo. In Autumn 1998, a rift within PRC opened between the two main exponents in the party, secretary Fausto Bertinotti and president Armando Cossutta. Disagreeing with the budgetary law proposed by the executive, Bertinotti wanted to withdraw the party's parliamentary support. Cossutta was instead strongly against such move. On October 9, this division resulted in a crucial parliamentary defeat for the centre-left cabinet, which definitively terminated. Cossutta resigned as PRC president and founded a new party called PdCI. A few days later, the newly born PdCI joined a new centre-left government, led by Massimo D'Alema, together with a number of other small parties, including the green FdV. Taking part in the new cabinet, PdCI and FdV demonstrated to highly value the presence in power of a centre-left cabinet in the context of a bipolar competition against Berlusconi's centre-right. Therefore, these parties were highly motivated to act as responsible junior partners. The communists felt this moral obligation more deeply as they had split from the rest of the PRC in order to keep on supporting the government. ‘The government had been in power not for long and we made a split from PRC to form it. Therefore, the willingness to look as responsible actors prevailed’, a PdCI MP admitted.Footnote 7 ‘I never had the feeling that they would leave the government. Not because they cared about the offices. But due to a matter of credibility and political culture’, a DS MP confirmed.Footnote 8 This does not mean that the Greens were unconcerned about the stability of the government per se. ‘Leaving the government would have led to new elections that the centre-left would have risked losing’, a green MP said.Footnote 9
These two parties were both in principle against the NATO military intervention in Kosovo and heavily contested Italian contribution to it, thereby demonstrating to attach a high extent of salience to the issue. Junior partners' contestation started even before OAF had begun. In fact, both PdCI and FdV immediately criticized Prodi government's decision to sign the NATO activation order on 12 October 1998, supporting a military intervention to solve the crisis and granting the use of Italian military bases to the allies. Cossutta defined it as ‘a mistake’.Footnote 10 In the following months, as the situation in the area worsened, the hypothesis of a NATO military intervention to protect the civilians became increasingly concrete, further exposing frictions in the new centre-left cabinet. In fact, while D'Alema reaffirmed Italy's loyalty to NATO,Footnote 11 Cossutta claimed that its party ‘was against the use of Italian military bases and a military intervention in Kosovo’.Footnote 12 At the eve of the intervention, FdV leader Luigi Manconi argued that divergencies existed within the cabinet and that its party disagreed with the prevailing ‘blind obedience towards the Atlantic alliance’.Footnote 13 In his memoirs, D'Alema himself acknowledged that, as the first former communist to lead an Italian cabinet, that he had to prove his loyalty towards the United StatesFootnote 14 . Predictably, the beginning of the air strikes on 24 March 1999 and the Italian decision to provide a substantial contribution nourished junior partners' criticism. A few days later, Cossutta and Manconi released a joint statement asking the executive to ‘publicly condemn’ the bombing campaign.Footnote 15 Notably, while PdCI cohesively opposed the NATO operation, FdV was more divided. In fact, following a humanitarian approach to foreign policy and the interventionist stance of its influent German counterpart, a fraction of the party supported the intervention as it aimed to protect the lives of civilians in Kosovo. ‘Some of the party activists were not against the NATO intervention’, a Green MP affirmed.Footnote 16 ‘Within the party, some argued that, when a humanitarian disaster is ongoing, just watching it happen is not the right thing to do’, another one added.Footnote 17 Lack of cohesion on the issue further decreased this party's propensity to defect.
Further highlighting the high extent of salience attributed to this issue, during 3 months of air strikes, PdCI and FdV were sincerely determined to steer the position of Italy towards a dovish direction. ‘Our goal was to push the Italian government to find a negotiated peace to the conflict and not let it end with a military victory. This was what we lobbied for’, a PdCI MP said.Footnote 18 ‘We insisted that the Italian government had to work for a restart of negotiations between the two parties’, another PdCI MP confirmed.Footnote 19 ‘We decided to stay in the government and try to make an effort to push the government to give priority to a diplomatic solution within the international community’ a Green MP stated.Footnote 20 Rumours regarding a possible ‘war cabinet’ between the centre-left and centre-right further encouraged junior partners not to defect.Footnote 21 Junior partners were afraid that such development would have pushed Italy towards a more hawkish position as centre-right parties were even ready to support a ground invasion.Footnote 22 ‘If we had left, a new government with the support of the centre-right would have formed. We were sure about that’,Footnote 23 a PdCI MP said. ‘Cossutta was in two minds between the belief that a new government with the centre-right could have formed and consistency with its ideological beliefs’ the then Minister of Defence, Carlo Scognamiglio, wrote in his memoirs on the Kosovo crisis.Footnote 24 ‘Within the party several positions coexisted about whether to stay in the government. In my opinion, the strongest argument for staying was that a fall of the government would have provoked a shift towards an even more aggressive stance’, a green MP said.Footnote 25
In order to steer the position of the cabinet towards a more dovish direction, thereby demonstrating to attach salience to the issue, these parties employed two instruments. First, they made use of parliament's monitoring and scrutinizing powers. For example, right after the beginning of the mission, PdCI and FdV called for a parliamentary debate on the issue. The more pacifist positions in the cabinet strongly affected the tone of the Prime Minister's speech and the wording of the governing parties' shared resolution. On April 26, referring before the Chamber of Deputies, D'Alema emphasized that the military forces' duties would be restricted to ‘the integrated defence’, meaning that Italian jets would not target Serbian cities. The resolution also ‘committed the executive to push NATO allies to stop the bombings and restart the diplomatic activities’.Footnote 26 On April 13, Parliament debated the issue once again while also approving the participation of Italy in the humanitarian mission Allied Harbour, aimed at providing aid and shelter for the refugees from Kosovo. Partly thanks to the efforts of PdCI and FdV, the shared motion explicitly denied any intervention on the ground.Footnote 27 In the last key parliamentary debate, on 19 May, also due to junior coalition partners' pressure, D'Alema indicated a road map for the suspension of the bombings.Footnote 28 Furthermore, junior partners extensively employed parliamentary questions to monitor the cabinet on the issue. For example, on May 4, Tullio Grimaldi, a PdCI MP, urged the executive to rule out that ‘the presence of Italian troops on the ground means a future deployment in a ground attack’.Footnote 29 With a marked environmentalist connotation, the greens repeatedly expressed concerns about the environmental damage of bombs discharged by warplanes on the sea. They also once called the government to ‘commit for a suspension of the bombings’.Footnote 30
Secondly, manifesting salience for the issue, junior partners increased their pressure on the government through threats and symbolic actions. For instance, PdCI threatened to withdraw its Ministers if the Italian government did not insist for a ceasefire during Easter.Footnote 31 In addition, both parties engaged in autonomous diplomatic missions to talk with the parties involved in the conflict. Using its contacts with other communist countries embarked in a diplomatic mission, Cossutta visited first Moscow and then Belgrade, concluding that ‘Milosevic was ready for a compromise’.Footnote 32 ‘We were in favour of the peace. We suggested that the government should keep the channels open Milosevic. Cossutta's travel, aimed at convincing Milosevic to stop military activities in Kosovo, was very symbolic in this sense’, a PdCI MP argued.Footnote 33 A few FdV exponents also went to the area to act as mediators. ‘Other members of the parliamentary group and I went on a mission Serbia and Albania to talk with the two parties in conflict’, a Green MP stated.Footnote 34
Libya
The events unfolding in Italian politics in the second half of 2010 are extremely relevant to analyse LN's decision to remain in the cabinet. On 30 July, the tension between Silvio Berlusconi and Gianfranco Fini, the leaders of the two parties that merged to form Popolo della Libertà (PdL), erupted in the departure of the latter, who established its own parliamentary group, called Futuro e Libertà (Fli). Consequently, the centre-right cabinet, led by Berlusconi, lost its parliamentary majority and had to rely on Fli's external support in order to survive. Despite passing a key confidence vote at the end of September, the collapse of the government seemed very near. LN was particularly pessimistic about the capacity of such a weak cabinet to be effective and started to call for new elections. On 17 October, LN's picturesque leader, Umberto Bossi, said that ‘if bills sometimes fail to pass, we have to go to ballots’.Footnote 35 ‘Lega Nord had already asked to go to new elections in the end of 2010, when Fini decided to defect from the government’, a PdL exponent revealed.Footnote 36 This can be interpreted as an evidence that they did not care much about the stability of the cabinet. Therefore, it seems that they did not fear to take the blame from their voters for bringing the government down.
The Libyan crisis, bursting in February 2011 and prompted by brutal repression of public protest by the regime, presented as a significantly thorny issue for the Italian government. Only a couple of years before, culminating a long strategy of rapprochement with the former colony, Berlusconi had signed a treaty of friendship with Libya. The treaty with Gaddafi mainly concerned Italian investments in the energetic sector and measures to contain the influx of illegal migrants from Libyan coasts to Italy. Given this, Berlusconi's reaction was extremely cautious. However, as attacks against civilians intensified and pressure from allies mounted, the Italian government was forced to take a more critical position against the dictator, adhering to sanctions against the regime. Notwithstanding Italy's diplomatic efforts, a Western military intervention in Libya became more likely day after day. As much as Berlusconi, LN disliked such a prospect. On 8 March, Roberto Maroni, Minister of Home Affairs and key LN exponent, warned about ‘not leaving the country in the hand of terrorists, unlike Afghanistan’ and asked for help from other European countries to share the burden of migrants according to a quota system.Footnote 37 In this statement, the opposition to the possible intervention in Libya was combined with concerns regarding the increase in illegal immigration towards Italy. Therefore, LN attributed salience to the issue as it was linked to another key issue in their agenda, that is, border controls. The interviews make this contradiction emerge clearly. On the one hand, an isolationist approach to the foreign policy comes up. ‘If you send a bomb, then it comes back at you. The terrorism that we experience is caused by our own past interventions. I would like to see no military interventions at all, unless they occur under well-defined circumstances’, an LN MP clearly stated.Footnote 38 On the other hand, another MP claimed that: ‘we had the issue of migrants and solving it was very important’.Footnote 39
Completing its half-hearted turnaround, Italy decided to support the no fly-zone on Libya, authorized by UN Security Council Resolution 1973, and provided the use of its strategic military bases for the military operation. A rift opens within the cabinet as LN openly criticized the mission led by the United States, United Kingdom and France and Italian government's position of support. Three days later, Bossi defined the intervention as an ‘Anglo-French plot to steal Italy the oil and gas and let it be invaded by millions of migrants escaping from the war’.Footnote 40 Therefore, condemnation for the military operation went hand in hand with the issue of migration and the competition for the control of energetic resources in Libya. The cabinet's decision to actively participate in a new NATO-led operation called ‘Unified Protector’ on 25 April raised the tone of LN's contestation. Bossi argued that ‘Italy was going to be invaded by millions of migrants’Footnote 41 and pointed out that the Italian contribution was too expensive and that it would have led to an increase in taxes. Salience was again placed on the domestic consequences of the intervention, rather than the intervention itself. The limited relevance attributed to opposition to intervention in Libya could also be inferred by the fact that, after a hypothetical date of the end of Italian contribution was set in May, LN stopped contesting. However, the air strikes continued until the end of October, when Gaddafi was killed by rebels on the ground. An LN MP explicitly summed up the lack of salience attributed to the issues of foreign and security policy. ‘The party was absent in foreign policy. Our leader rarely spoke about it’,Footnote 42 he admitted.
LN instead made considerable efforts to steer the government's decision towards a more dovish direction, through the use of parliamentary resolutions, threats and symbolic actions. On 19 March, the junior partner deserted the meeting of the parliament's foreign affairs committee in which Franco Frattini, the Minister of Foreign Affairs, announced the decision to concede the military bases to the allies. An LN interviewee (n.12) confirmed that it was conceived as a protest. Moreover, right after the decision to implement a no-fly zone, LN demanded a parliamentary debate on Italy's participation in the multilateral intervention against Gaddafi. The government conceded it on 24 March. The resolution drafted by coalition parties heavily reflected LN's demands as it advocated sea patrolling against human trafficking in the Mediterranean Sea and the concept of burden-sharing among European countries on the issue of migration.Footnote 43 Furthermore, on 3 May, the junior partner presented its own motion in Parliament on the intervention in Libya. In the document, other than stressing their concerns regarding the costs of the air strikes and the potential increase in the migration flows, they asked the government to communicate a precise date in which Italy's involvement in the conflict would end. Bossi said that Berlusconi and his party had to sign it ‘if they wanted the government to stay in power’.Footnote 44 Eventually, the resolution passed thanks to the votes of both the junior and the senior partner.
However, such efforts seemed more directed to appease their own voters rather than based on a genuine interest for the issue. ‘The absence in the vote in the committees was a strategy to please their electorate’, a PdL representative argued.Footnote 45 ‘The resolutions in parliament were balancing acts not to let the government fall. Because we did not want this to happen’,Footnote 46 an LN MP admitted. Moreover, in contrast to extreme junior partners' role in Kosovo, they did not monitor and scrutinize the conduct of the operation by formulating any parliamentary question during the intervention.
Discussion and conclusions
Against a background of a disliked foreign policy outcome, junior partners face a dilemma: staying in the cabinet or leaving? This choice has substantial implications on their capacity to influence policymaking and may considerably impact on their future results at the polls. While a few works hint that conflict on foreign policy does not usually trigger government's termination (Kaarbo, Reference Kaarbo2012; Coticchia and Davidson, Reference Coticchia and Davidson2018), little effort has been done to address the causal mechanisms underlying such outcome. In this paper, I tackled this question: why do junior partners stay in the government in spite of a foreign policy decision distant from their preferences?
In order to provide an explanation, in the context of decisions regarding the participation in military interventions, I hypothesized two possible mechanisms: lack of salience attributed to the issue (H1) and fear of being blamed by voters as an irresponsible coalition partner (H2). Through the use of qualitative data, including original interviews, these hypotheses were tested in two cases: Italy's decisions to participate in OAF (Kosovo, 1999) and OUP (Libya, 2011). The cases share similarities in terms of scope of the mission and outcome, with junior partners deciding to remain in the cabinet, notwithstanding their criticism towards the intervention. However, they fundamentally differ for ideological leaning of the junior partners: extreme-left in the case of Kosovo and far-right in the case of Libya.
Interestingly, the mechanisms played out differently according to the position of the junior partner on the left-right axis. In the case of Kosovo, in line with H2, the communist PdCI and the green FdV were particularly concerned of being blamed as irresponsible by their own voters for paving the way for the centre-right's return to power. In particular, this mechanism was even more decisive in explaining PdCI's permanence as the survival of a centre-left government constituted its own raison d’être. In contrast, both parties considered Italian participation in NATO-led bombing campaign as a highly salient issue as demonstrated by persistent contestation and the employment of various strategies to put pressure on the cabinet and steer its position towards a more dovish direction. In the case of Libya, the far right and autonomist LN certainly did not regard the participation in the intervention as a salient issue per se as it was rather concerned by its potential side effects on immigration and tax increases at home. Confirming H1, this variable was the main driver behind the decision to remain in the government, with little evidence for other mechanisms. For instance, fear of being labelled as an irresponsible coalition partner was totally absent as evidences suggest that LN threatened to make the government collapse at a previous stage. Table 1 resumes the empirical findings in the two cases.
Presence of explanatory factors in the two cases
Through these findings, this article primarily contributes to the understanding of the relationship between political parties and MOAs in Italy during the ‘Second Republic’. In fact, it explains why coalition cabinets composed by parties with supposedly diverging point of views on this issue managed to survive. Furthermore, in line with Coticchia and Vignoli (Reference Coticchia and Vignoli2018), the article highlights variation in extreme partners' opposition to MOA in Italy: more ‘ideological’ for extreme-left parties and more instrumental for LN.
Moreover, the article also enriches the broader comparative research agenda on the party politics of foreign policy. First, it introduces an innovative research puzzle: ‘why do junior partners stay in the government in spite of an undesired foreign policy outcome?’. Addressing this question is extremely relevant to explore the junior partner's patterns of influence. In fact, a credible threat of leaving the government increases junior partner's hijacking power on the senior partner and, by the same token, its impact on foreign policy outcomes. In addition, notwithstanding the peculiarities of the Italian party system and the extreme ideological leaning of the parties taken into account, these findings could be extended to other cases of small and middle powers in Western Europe, frequently governed by coalition cabinets, such as Netherlands and Belgium.
Finally, the article has substantial implications for the study of coalition politics and government termination. Junior partner's lack of influence in both cases challenges the application of veto player theory on coalition foreign policymaking. This may suggest that such a theoretical perspective may be not effective on low salience issues like involvement in multilateral interventions. Moreover, the article provides interesting insights concerning the impact of coalition conflicts on government's termination, in Italy and abroad. The literature on government termination has mostly quantitatively identified variables making a cabinet more likely to collapse (Laver, Reference Laver2003). These studies may be complemented by further qualitative works investigating reasons underlying junior partners' permanence in the cabinet.
Funding
The research received no grants from the public, commercial or non-profit funding agency.
Acknowledgements
The author would like to thank all the participants to the panel ‘political parties and foreign policy at the time of populism’, that took place at 2019 SISP Conference, for the insightful suggestions on an earlier version of the article and the two anonymous reviewers for the useful comments.
Appendix
Table A (Footnote 5). Composition of Berlusconi IV government at the beginning of the legislature (May 2008).
Table B (Footnote 5). Composition of D'Alema I government at the beginning of the legislature (October 1998).
Table C (Footnote 7). Overview of the interviews.
Examples of questions asked during the interviews (Footnote 7)
Operation Allied Force (Kosovo, 2009), interviewee 6, Federazione dei Verdi MP
• Before falling, on 12 October 1998, the Prodi cabinet, signed the NATO activation order, warning the members of the organization of a possible intervention in the Kosovo crisis. At the moment of joining the D'Alema cabinet, were the Greens aware of the eventuality that Italy would support such military operation?
• The green party was critical of the government's decision to concede Italian bases to NATO at first and, subsequently, of the substantial involvement of the air forces in the strikes. Why did the party decide to stay in the cabinet, nevertheless? Which factors pushed the Greens not to defect?
• Did disagreement regarding the decision to remain at the government exist within the party?
• Which efforts did Federazione dei Verdi make to steer the cabinet's position in a more dovish direction?
• Another junior coalition partner, Partito dei Comunisti Italiani (PdCI), was also strongly against the intervention. PdCI decided not to defect too. Why do you think they took such decision? Do you think that their reason to stay differed with respect to the one adopted by your party?
Operation Unified Protector (Libya, 2011), interviewee 13, Lega Nord MP.
• In 2011, Lega Nord expressed its strong aversion towards the multilateral intervention in the Libyan crisis and the government's decision to make a contribution to it. Why?
• Notwithstanding this dislike for the military operation, Lega Nord decided to remain in the cabinet. Why?
• Did disagreements regarding the decision to remain in the government exist within the party?
• Which efforts did Lega Nord make to steer the cabinet's position in a more dovish direction?
