SISP Conference 2022

Since 2015, the Spanish party Podemos has developed a new form of campaign financing: microcredit. By using digital technologies, microcredits are electoral funding initiatives based on civil loans from the party’s activists and sympathisers (at zero percent interest rate) that the party returns to the lenders once it receives the State subsidies for elections, normally one year later. This way of financing has been used ever since, albeit the current Spanish electoral and financing regulation does not specifically contemplate this type of campaign financing. Other parties have also started microcredits campaigns, including smaller left parties (such as United Left, ERC, Más País) and, notably, the Socialist party (PSOE). However, unlike Podemos, which limits microcredits only for campaign funding, the Socialists have implemented them with a different perspective, as a complementary tool for financing electoral campaigns and granting a very competitive interest rate. It has also broadened the scope of this financial innovation, including other party’s processes that are not linked to electoral campaigns. Accordingly, the PSOE uses microcredits for raising funds for financing the party’s Plan for Ecological and Digital Transition, that aims to reduce the party’s CO2 imprint, foster its digital transformation of all its internal processes, reducing the environmental impact of the party’s public events or the sustainability of the party’s local, regional and national headquarters. Thus, we can distinguish two types of microcredits, with different rationale, objectives, uses, and way of communicating to members and sympathisers. Following the distinction between the two different types of online crowdfunding, we can distinguish between symbolic or equity microcredits. The first type refers to Podemos’ microcredits, as lenders do not receive any financial benefit and lend their money as a way for supporting the party, its ideology, and policies. In a certain way, due to specific characteristic of how this financial mechanism is framed and communicated, we could also consider it as an ideological or even a populist microcredit. On the other hand, the equity microcredit implies a financial return, with a very competitive and quite save investment. Contrary to the other type, this financial tool is framed as another option in the mix of different financial mechanisms at the party’s disposal for financing different party actions, not necessarily related to elections. Furthermore, the two types of microcredits also imply a diverse relationship between the party and its members and sympathisers. What are the characteristics of these two microcredits? Why have they been implemented? What are the consequences? This paper aims to analyse the logic, implementation, and uses of microcredit by comparing both parties and discussing the possible consequences for the relationship between members and the party through party financing. We do it by using publicly available data on microcredits, monitoring the communication of each microcredit campaign by recollecting the emails and communications sent by each party, and with interviews with the main party cadres in charge of the implementation of the tool.

Prior to 2000, the UK operated one of the most liberal political finance regimes of any established democracy. Parties were highly dependent on private financing, state funding was minimal, limited transparency requirements existed with respect to party income or expenditure, and there were no limits on national election spending. Far-reaching reforms introduced by Labour in 2000 changed this regulatory environment radically, with the aim of levelling the playing field in party funding and election spending. However, Labour’s reforms did not challenge the assumption that parties should predominantly be funded through private means – a decision which had profound, unforeseen consequences. In this paper, we show how the Conservatives successfully adapted their fundraising strategies in light of Labour’s reforms, enabling them to greatly outspend Labour at four general elections from 2010-19. Using the public registers created by Labour’s reforms, we document two key features of this Conservative adaptation to the new party finance regime. First, we show that the party has shifted from its previous reliance on mainstream corporate donations to a mix of donations from wealthy individuals and new corporate entities, such as hedge funds, technology firms and new media enterprises. Second, we highlight how Conservatives donors have made use of a range of mechanisms to split donations from a single source among multiple donors or by creating organisation entities for the purpose of aggregating donations from multiple source. These legal techniques of “fame avoidance” are likely to have been crucial to the party’s continued success in generating such large sums through private donations.

Online fundraising has remained for a long time virtually absent from party websites in continental European countries. This can be explained by the importance of state funding to political parties across the region. Indeed, research has shown that political parties are more actively engaged in fundraising activities in countries where private sources of income are dominant, such as in the Anglo-Saxon area. This also holds for the online dimension of political fundraising. This paper observes the over-time evolution of online fundraising by political parties in Italy, a country where direct public funding to political parties was recently repealed. Thus, from being among the countries where parties mostly depended on state funds, parties in Italy were forced to turn to alternative sources of income. The Italian case offers an excellent context where to observe the way in which parties adapt to new forms of financing in a growing digital environment. By means of a content analysis of the functionality and design of party websites adapted from coding schemes originally developed by Gibson and Ward and of secondary literature in the field, the analysis will compare the evolution of fundraising activities (2000-2022) by political parties on their websites and observe if and how online fundraising patterns vary overtime and across parties. While empirically focusing on the Italian case, the paper will also provide a comparative discussion on the subject. Indeed, especially following the 2008 financial crisis, various are the European countries that amended laws to reduce the subsidies available to parties, suggesting a diminished state intervention in party financial matters in the coming future. It is therefore reasonable to expect that parties throughout the European continent will adapt and invest in innovative tools for raising money, with the potential added benefit of them reaching out more to citizens.

This paper aims to assess the causal impact of the end of the Cold War on party funding. It is well-known that in Italy, the end of the Cold War contributed towards the transformation of the party system in the 1990s. However, there were possibly more subtle effects of the Cold War. To analyse these effects, the paper engages in a process-tracing of party funding reforms in Germany. Particular emphasis is paid to the substantial changes in private funding to political parties. Prior to 1994, private donations were tax-deductible up to DM 100.000, which constituted an enormous incentive to donate to political parties, especially for corporate business from which most donations stemmed. However, the anti-establishment Green Party challenged this party funding regime and took a number of its regulations to the Constitutional Court in the 1980s. Remarkably, the Court initially only called for minor changes but then ruled that the whole regime of party funding required substantial reform in 1992. This led to major changes during which the tax incentives for donations disappeared. Corporate donations ceased to be tax-deductible after 1994. The paper hypothesizes that these changes were facilitated by the end of the Cold War logic. Whereas the international bloc confrontation prior to 1990 served as an incentive to enhance partisan unity within nation states of “the” West, such incentives disappeared after 1990 and gave way to more emphasis being paid to internal problems, most notably, political corruption. This led to different perspectives on the funding of political parties, especially from private donors: During the Cold War, the major goal of party funding was to ensure that (allegedly) extremist parties were cut off financial resources. After the end of the Cold War, notions equal opportunities between parties, especially between established and new ones, became more important. What previously appeared as a cordon sanitaire to fend off radical newcomers was now regarded as either a case of or an opportunity for political corruption. This is how the end of the Cold War affected party funding reform in Germany, especially with respect to private funding. The most important effect of the 1994 reform (and, accordingly, the end of the Cold War) was the virtual disappearance of large donations which increasingly substituted by membership dues (with ever larger individual amounts from ever less members). In order to assess the impact of the Cold War on these changes to party funding in Germany, the paper engages in a process-tracing analysis of all reform attempts of the Green Party both in the Bundestag and at the Constitutional Court between 1983 and 1994 based on parliamentary papers, newspaper articles, and Court rulings. It is expected that the perception of corruption becomes increasingly negative over time whereas the challenge posed by new parties is regarded less and less as a problem for democracy.

Political parties must be legitimate in order to operate as agents of citizens' interests. Historically, the political legitimacy of mass party organizations was heavily dependent on their large membership base. As is generally known, since the mid-1960s, the organizational model has collapsed, becoming more state-oriented than citizen-oriented (Kirchheimer 1966). (Katz and Mair 1995). In times of political instability, membership and party loyalty have fallen, leading to the rise of alternative forms of partisanship and civil society representation (see Scarrow, Poguntke, and Webb 2017, Bolleyer 2020). This paper looks at partisanship and civil society interests from a different perspective: the financial one. Financial resources have long been considered for their perceptive power and the public dependence of parties, particularly in European democracies. Private contributors, on the other hand, will help us comprehend the legitimacy foundation of the public political actor. Recent changes in the way politics is structured have resulted in digitalization as a kind of modernization for both old and new political players. This paper focuses on the adaptation process and analyses how political parties in Western countries connect to the web society and their digital fundraising strategies. Taking into account the Five Stars Movement (FSM) as an example of a digital party, this work aims to understand the financial dynamics of a “challenger” party that constantly remarks on its differences with traditional political organizations. The analysis of the financial resources of FSM contributes to understanding its electoral success and the problem it is facing to institutionalize. The new fundraising tactics pose serious concerns about the legitimacy of political parties and the amount of transparency that they enjoy now.

In 2003, a separate finance regime for political parties at European level – also called “Europarties” – was established. These rules not only introduced public funding for these European political parties, but also set specific conditions and requirements regarding private funding. On the one hand, the use of donations has been clearly regulated by the funding rules. On the other hand, the design of the finance regime has stimulated the Europarties to make as much use of donations as possible. This paper will examine the incentives and constraints of the regulatory framework for Europarties with regard to donations, the pattern and evolution of donations over time and across parties, their fundraising strategies and the motivation of donors to financially contribute to these parties. The analysis is based on legal texts, public and internal documents of the European political parties and the EU institutions (obtained, among others, through transparency requests and a collaboration with research journalists), complemented with semi-structured interviews. The findings will show how the Europarties have conducted very different practices, with some banning donations altogether, and others seeking donations of private companies in exchange for contracts.

The study of party membership is a traditional sub-field of literature on political parties, especially studies based on internal party organization. Among other things, party members' research focused on why people join parties and how to increase their members' opportunities for participation (van Haute and Gauja 2015; Demker et al. 2019; Bale et al. 2019). The decline of party membership (i.e. Dalton and Wattenberg 2002) has contributed enormously to the growth of these studies. This paper will investigate the relationship between the party and its members by focusing on two organizational innovations introduced by the Italian Democratic Party: a) the selection of candidates through open primaries; b) the organization of spaces for discussion opened to party members and sympathizers (so-called Agorà). What opinion do the party members have of these participation tools? What factors determine the different views of the party members? Is it possible to build a typology of party members based on their attitude towards these tools? Employing an original dataset from a survey administered to Italian Democratic Party members between January and February 2022 (about 4,000 respondents), this work answers these questions.

Le modalità di campaigning elettorale sul campo sono da sempre influenzate dall’evoluzione delle tecnologie di comunicazione. Se, da un lato, l’invenzione della televisione segna il passaggio a forme sempre più deterritorializzate e mediate di campaigning, l’adozione dei computer per lo stoccaggio di dati e la facilitazione delle attività di campaignin ha facilitato un ritorno di attenzione per il campaigning locale. Le tecnologie digitali sono state utilizzate per velocizzare e semplificare alcune attività, in un contesto in cui la centralizzazione strategica conviveva col tentativo di utilizzare le nuove capacità di data storage per raffinare le modalità di targeting delle comunità locali, specialmente in sistemi first past the post. In un contesto in cui il ritorno sull’investimento del campaigning sui mass-media è decrescente per via di una progressiva saturazione si fa inoltre spazio l’idea che un maggiore focus sui seggi chiave, assistito dalla mobilitazione di volontari locali capaci di modulare con efficacia il messaggio delle campagne nel constesto specifico, possa garantire un vantaggio elettorale importante. Con le campagne di Howard Dean ed Obama negli USA si assiste ad un salto di qualità in questa traiettoria, che si prolunga nel decennio successivo e sbarca in UK con le campagne del Labour nel 2017 e 2019. Per la prima volta le nuove tecnologie digitali vengono utilizzate per soppiantare parte dell’infrastruttura organizzativa tramite l’adozione di piattaforme digitali alle quali viene delegato il compito di coordinare le attività di voter persuasion svolte da volontari. Per usare un’espressione coniata da Bennet e Segerberg: all’organizzazione si sostituisce la comunicazione. Si sviluppano cioè modalità di campaigning connettive e non più collettive, in cui le affordances decentralizzanti e cost-lowering della rete vengono utilizzate per coordinare grandi quantità di volontari non affiliati ai network locali di partito e spesso neanche ai partiti tout court. Inoltre la creazione di app, che assumono di norma la forma di mappe presso le quali è possibile postare e trovare eventi e che spesso integrano altre funzioni come la creazione di gruppi Whatsapp per il coordinamento volontari e sezioni dedicata alla data entry e il download delle liste di contatti, permette di eliminare almeno in parte i livelli intermedi di management, come accade nel caso dei digital parties. La generale facilità di accesso ed apertura delle app, che permette anche a chi non è inserito nelle reti di partito di partecipare, permette infine una significativa personalizzazione dell’esperienza di partecipazione: se si eccettuano alcuni “super-volontari” i partecipanti non sono in alcun modo vincolati a partecipare a specifiche attività, potendo scegliere se, quando e con quale coinvolgimento aderire alle attività di volta in volta grazie alla facilità di accesso e alla grande visibilità delle informazioni sulle attività sulle app. Questa sostituzione del management verticale attraverso la facilitazione della partecipazione tramite piattaforme digitali ha provocato parecchio ottimismo in diversi commentatori. In particolare, la semplificazione delle strutture organizzative delle campagne è parsa una potenziale arma nelle mani di candidati outsider. In più, il filone di dibattito sulle tecnologie digitali informato da un certo ottimismo tecnologico ha visto in queste forme di campainging una potenziale innovazione democratica, capace di porre direttamente nelle mani dei cittadini il pieno potenziale decentralizzante delle ICTs. Come però mostra Stromer-Galley questo tecno-ottimismo si scontra con una realtà molto diversa, in cui il controllo sulle piattaforme rimane saldamente nelle mani dei campaign leaders che manipolano le affordances delle stesse nonché il livello di accesso ai dati per il targeting al fine di mantenere un controllo sulle interazioni ed indirizzare le attività dei volontari verso gli obiettivi della campagna. Su questo punto, però, le affermazioni di Stromer-Galley sono imprecise; l’autrice parla infatti di “controllo dell’interattività”, facendo riferimento al fatto che le campagne permettono interazioni decentrate solo quando ciò è in linea con la facilitazione dello sfruttamento del lavoro gratuito dei volontari e non quando ciò permette un controllo bottom-up sulla campagna. In questo caso, occorrerebbe un focus più specifico, facilitato dalle categorie introdotte da Flanagin e colleghi. Non è infatti l’interattività ad essere limitata dalle piattaforme; la comunicazione tra volontari è in realtà fortemente incentivata, specialmente se si compara il connective campaigning con le precedenti forme. Il controllo sulle campagne è piuttosto ottenuto tramite la riduzione delle possibilità di engagement, ovvero di coinvolgimento dei partecipanti nella strutturazione dell’organizzazione dell’azione collettiva, che in questo caso si traduce nell’impossibilità di discutere ed influenzare il design delle piattaforme. É nella limitazione delle possibilità di engagement con l’organizzazione che viene neutralizzato il potenziale decentralizzante delle ICTs. Come nei digital parties, la sostituzione degli strati intermedi dell’organizzazione attraverso le piattaforme non è adeguatamente compensata attraverso nuove forme di linkage e feedback. Ma la questione non si esaurisce qui. Sicuramente le infrastrutture digitali sono fondamentali per mantenere la coerenza organizzativa, ma non sono l'unico elemento a limitare interattività ed engagement; oltre che su queste occorre continuare a concentrarsi su quelle che portemmo definire infrastrutture umane, ovvero le reti di relazioni sociali, sia formalizzate che informali, e le norme e le aspettative che definiscono l'uso collettivo degli strumenti che compongono l'infrastruttura digitale delle campagne. In parte il concetto di infrastruttura umana deriva dall'uso che Tufecki fa del concetto di infrastruttura, distinguendo tra mobilitazioni tradizionali e digitali crowd-enabled: mentre nel primo caso, secondo l'autrice, le operazioni di logistica e coordinamento dell’azione richiedevano una solida infrastruttura organizzativa preesistente le mobilitazioni, gli strumenti digitali consentono di esternalizzare il lavoro di coordinamento a piattaforme che consentono una comunicazione più facile e veloce tra i partecipanti atomizzati. Dal momento che Tufecki utilizza il concetto di infrastruttura per identificare il lavoro di costruzione delle precondizioni organizzative dell'azione ne deriva che in questo framework l'infrastruttura umana non è più presente nelle mobilitazioni digitali, soppiantata dall'infrastruttura digitale. Tuttavia questo approccio non è adeguato al caso delle campagne connective. Le campagne digitali infatti prevedono una commistione di logiche connettive e collettive, tra cui varie forme di coordinamento diretto ed esplicito e diversi tipi di alleanze e accordi tra le diverse componenti delle campagne. Inoltre, è importante ricordare che le campagne digitali richiedono spesso una quantità di lavoro di back-office che per molti aspetti può essere riconducibile al concetto di infrastruttura proposto da Tufekci. L’ammontare di back-office necessario è in larga parte inversamente proporzionale al livello di decentramento delle mansioni di data entry data e management: maggiore il livello di accesso concesso ai volontari, tanto meno back office sarà necessario. Come già accennato, tuttavia, le infrastrutture umane non identificano semplicemente le strutture organizzative in cui è incorporato l'uso degli strumenti digitali. Anche le aspettative collettive definite dalle rappresentazioni condivise rispetto ai comportamenti adeguati e ai ruoli legittimi rispetto alla situazione della campagna hanno un forte impatto sul mantenimento della coerenza organizzativa. Gli organizzatori della campagna devono fare in modo di essere riconosciuti come attori competenti e legittimati a guidare l’azione e utilizzare questo capitale simbolico per spingere i volontari a concentrarsi solo sulla vittoria piuttosto che a mettere in discussione le priorità e la struttura della campagna; ciò è necessario per garantire il controllo sulle attività anche quando il funzionamento delle infrastrutture digitali e la formalizzazione dei ruoli organizzativi non offrono un vantaggio naturale e diretto sui partecipanti. In tutti questi casi, gli organizzatori devono convincere i volontari a subordinarsi agli obiettivi della campagna. Questo perché, anche se l'interattività e l’engagement sono manipolate attraverso la progettazione delle piattaforme digitali, c'è una componente ineliminabile di decentramento che apre margini all’uso degli strumenti digitali da parte dei volontari per auto-organizzarsi al di fuori del controllo dei leader della campagna. Come abbiamo visto, la caratteristica principale delle campagne connective è l'uso delle tecnologie digitali per aumentare la quantità di lavoro volontario utilizzato nelle mobilitazioni pre-elettorali. Ne consegue che la forma specifica assunta da ciascuna campagna dipende essenzialmente da due fattori: il design degli strumenti digitali e l’organizzazione del lavoro umano. Naturalmente, i due fattori interagiscono: da un lato, le tariffe delle piattaforme digitali consentono o precludono determinate azioni, incanalando le attività dei volontari nella direzione desiderata dai leader politici e organizzativi delle campagne; d'altra parte, le persone possono utilizzare gli strumenti messi a loro disposizione in modi diversi. Per questo motivo, è necessario distinguere tra l'infrastruttura digitale delle campagne e la loro infrastruttura umana. Scomporre le campagne secondo questi due fattori può fornire una maggiore comprensione dei funzionamenti specifici di ciascun caso e consentirci di riconoscere diversi modelli organizzativi, cercando di districare gli effetti di ciascuno strumento e forma organizzativa.

Con all’orizzonte un turno di legislative che dirà se il secondo mandato presidenziale di Emmanuel Macron potrà o meno affidarsi al supporto di un’adeguata maggioranza rappresentativa, il partito di Macron, nato come un’impresa individuale e come nuovo tipo di partito personale, sembra aver cambiato formule e soluzioni - da En Marche a La République En Marche e fino alla sperimentazione della coalizione di Ensemble - mantenendo tuttavia ferma l’essenza di una sua innovazione tipicamente post-democratica: dare espressione organizzativa all'individualismo tecnocratico. A qualche anno dalla sua genesi, tra successi e qualche flessione soprattutto inter-elettorale e di mid-term, come è cambiato il partito ‘di Macron’? La particolare struttura decisionale e di reclutamento del management politico che l’aveva caratterizzata dal principio com’è mutata, soprattutto dopo la prima esperienza di governo? Geografia e sociologia di supporters “militanti” e marcheurs sono mutate nel tempo? E con esse, anche le soluzioni organizzative proposte, considerando soprattutto la sfida della rielezione? Sviluppando un frame analitico ed empirico che copre gli anni dal 2016 al 2022, il paper intende provare a rispondere a questi interrogativi.

Digital technologies are offering a novel environment for political activities and, more specifically, for interactions between citizens and political actors. The literature on these topics is burgeoning (Barbera et al. 2021; Dommett et al. 2021a; Gerbaudo 2018). On the one hand, studies of political communication emphasize the changing nature of election campaigns and the reshaped relationship between leaders and supporters (Schäfer 2021; Dommett and Temple 2018; Chadwick and Stromer-Galley 2016), while on the other hand the literature on political parties examines in more detail the organizational implications of such digital shift (Dommett et al. 2021b; Dommett et al. 2021c). Against this backdrop, the paper intends to investigate the opinions and participatory attitudes of party members towards the new digital participation opportunities that their party organizations provide. To do so, we rely on original individual survey data. More specifically, we will use data derived from a survey administered to Democratic Party members in Italy at the beginning of 2022 (about 4,000 respondents). Precisely, we aim at identifying party members profiles according to their (degree of) digital activities, by controlling for variables such as: length of the membership, levels of intra-party activism, evaluation of intra-party democracy. Moreover, we will investigate the changing relationship between members and their party organizations in the new digital ecosystem.

Il carattere inclusivo del momento elettorale (di tutti gli adulti) è uno dei criteri fondamentali del processo democratico moderno fondato sul suffragio universale (Dahl 2000, 37-8). Tuttavia, esiste sempre una differenza più o meno ampia tra il numero di individui teoricamente titolati a votare sulla base dell’età (VAP, voting-age population) e il numero di quelli che lo possono effettivamente fare (VEP, voting-eligible population), ossia che sono autorizzati per legge a prendere parte alle elezioni. Sebbene risulti inferiore al caso statunitense (Mc Donald 2001), dove a dissuadere dal voto alcune fasce sociali si aggiungono le procedure obbligatorie di registrazione (Avery e Peffley 2005; Delwin 2013), il divario tra VAP e VEP si riscontra anche in Europa, con significative differenze tra i paesi e in base al tipo di categoria considerata (Geys 2006; Tuorto 2022). Il presente contributo intende mettere a confronto i paesi dell’area UE-27 (+ Gran Bretagna) rispetto ai criteri che regolano l’inclusione/esclusione elettorale dei gruppi sociali più suscettibili ad essere sottoposti a limitazioni transitorie o definitive nella loro partecipazione al voto, in relazione a specifiche condizioni e/o condotte pubbliche. In particolare, l’analisi si concentra sulle seguenti categorie sociali: persone con disabilità psico-fisica, persone che hanno commesso reati, cittadini non residenti (autoctoni all’estero) e residenti non cittadini (immigrati). L’eterogeneità di situazioni riscontrate all’interno della stessa area comune europea segnala, da un lato, le differenze esistenti sul piano legislativo e culturale (le rappresentazioni sociali sulla disabilità, la criminalità, la condizioni di straniero e la traduzione normativa di queste rappresentazioni). Dall’altro, l’allontanamento o esclusione sistematica di determinate fasce sociali dalle elezioni pone un problema di legittimità del momento per eccellenza di esercizio della democrazia, in presenza di barriere legali, amministrative, simboliche che rendono i risultati elettorali imperfetti e almeno parzialmente distorti.

International election observation has long been scrutinized in the past two decades by scholars trying to explain its aim, scope, and consequences. Being now a standard practice in democratizing countries, scholarly literature has almost reached a consensus on the ability of observers to deter election-day fraud, one of the most positive unintended consequences of election observation. This article deepens even further our knowledge through a natural experiment on polling-station-level election results. Using data from the Ukraine 2004 presidential election, it will show that observation is most effective in the counting phase, having both immediate and lasting effect on domestic political actors’ behavior. Results do support previous studies showing the usefulness of election observation in reducing election-day fraud but add an important finding that can improve the (unintended but welcome) observation effect.

Over the past few decades, almost all European countries recorded decreased levels of turnout. The literature indicated that much of this downward trend is driven by the abstentionism of young people. Previous studies investigated low youth turnout emphasizing context-and generation-based explanations, showing that younger generations vote less compared to older ones because of their low level of external efficacy and perceived lack of political representation. However, there is still lack of clarity as for how to interpret the detachment of younger people from the old politics of voting. No study has so far analysed the connection between the young people declining turnout and the actual representativeness of the political system. Taking the literature on turnout generational change as a point of departure, this study makes a step forward to examining the relationship between youth and electoral participation, shedding light on the impact of different types of representation, i.e. descriptive and substantive, on turnout. It does so by relying on a large-n database comprised of 47 legislatures and 27 European countries. The study’s findings demonstrate that when younger generations are descriptively and substantively represented, the generational gap in turnout is significantly reduced.

Studies on the QoE regard mainly countries with hybrid regimes, new democracies, or regimes in transition. On the other hand, research into established democratic regimes is much rarer. There are, however, two exceptions: the United States and, to a much more marginal extent, the United Kingdom (Norris, 2014; Clark, 2014a, 2015). The examination of the Italian case has first of all the purpose of filling a gap in the literature and expanding the studies on the quality of elections among the consolidated democracies. Italy, however, has some peculiar characteristics: a hierarchical administration of the electoral process, based essentially on top-down procedures, led by the Ministry of the Interior; considerable variability in the electoral systems used at different levels of government, both in a synchronic and diachronic sense; a generally higher level of perceived corruption than in most established democracies. The paper will be focused on the clarification of the concepts related to QoE and on QoE’s measurements. On the first side, the concepts of electoral integrity, electoral administration, and electoral malpractice will be discussed. On the second side, we will focus on the possible ways of measuring QoE, examining both measurements based on the subjective perception of the level of QoE (opinion polls among citizens and practitioners; expert surveys) and those based on objective information (i.e., residual voting; contested votes; complaints about electoral fraud). The paper will also apply one of the measures of the level of QoE to the Italian case. Following the Pew, in partnership with the Massachusetts Institute of Technology, QoE will be measured and explained by using the level of residual voting (Flavin and Shufeldt 2019) as a dependent variable. The analysis includes all the Italian parliamentary elections from 1948 to 2018 and is conducted at the provincial level.

In Italia, la gestione della pandemia è stata costellata da un aperto conflitto tra attori politici e istituzionali collocati nei diversi livelli territoriali di governo (Governo, Regioni, Comuni). Soprattutto nella prima fase dell’emergenza, questo conflitto ha generato un tipico blame game che è andato rapidamente a installarsi sulla frattura Nord/Sud. Rispetto a questa frattura, il virus ha capovolto il consueto scenario. Nonostante il peggiore rendimento politico-istituzionale, le regioni meridionali sono riuscite inizialmente a contenere i contagi. Al contrario, il moderno, efficiente e ben amministrato Nord ha fatto registrare in alcune sue aree una delle più alte concentrazioni di vittime di Covid-19 al mondo. Questa anomalia è lentamente rientrata nella successiva stagione pandemica quando anche il Mezzogiorno è stato investito appieno dall’onda pandemica ed è emerso, più in generale, il difficile rapporto tra regolazione centralizzata degli interventi e risposte locali non sempre coerenti con l’impianto nazionale. Con questo contributo intendiamo esplorare come è stata vissuta e giudicata questa contrapposizione istituzionale che ha fatto leva, nemmeno tanto velatamente, sui sentimenti campanilistici più che su quelli di natura politica. Per valutare se, in questa particolare situazione emergenziale, le appartenenze politiche abbiano prevalso sui sentimenti di appartenenza territoriale nel dar forma alle opinioni dei cittadini sull’operato del Governo e delle Regioni, prenderemo in considerazione le variabili relative alla regione di residenza, ai giudizi sull’operato del Governo e del Presidente della propria regione, all’autocollocazione sulla dimensione sinistra-destra e la propensione a votare per un determinato partito. La dimensione temporale delle rilevazioni, strutturate in quattro ondate che coprono il periodo aprile 2020-dicembre 2021, ci consentirà di valutare l’evoluzione dei giudizi in relazione sia all’andamento della pandemia (che ha colpito progressivamente tutto il paese, quindi anche le regioni del Sud) sia agli sviluppi del quadro politico nazionale (con la sostituzione del governo Conte con quello a guida Draghi) e regionale (con l’elezione di nuovi governatori in piena emergenza pandemia).

Nella fase della pandemia e soprattutto dopo la costituzione del governo Draghi sono cambiati in modo significativo i consensi per i partiti italiani, in particolare le intenzioni di voto per il M5s. L’insoddisfazione e le critiche nei confronti dei principali partiti erano molto cresciute dopo l’esperienza del governo Monti. Il M5s aveva raccolto e alimentato la protesta popolare, entrando per la prima volta in Parlamento con un consenso elettorale del 25%, prima forza politica nel nostro paese. Un successo mai riscontrato in tutta la storia politica italiana. Una sorta di rivoluzione che concludeva un ventennio di sostanziale equilibrio bipolare, con la costante alternanza – da una elezione all’altra, dal 1994 al 2008 – tra governi di centrodestra e di centrosinistra. Dopo quel primo grande exploit, nel 2018 la performance del M5s come noto migliorò ancora, arrivando quasi al 33% dei voti. Combinando in maniera inedita un po’ di populismo e un po’ di innovazione tecnologica, era diventato il punto di riferimento di elettori stanchi della politica tradizionale e delle antiche contrapposizioni ideologiche tra destra e sinistra. Il Movimento fondato da Beppe Grillo aveva ottenuto in quelle due elezioni consensi provenienti da aree politiche molto diverse, diventando la formazione con la rappresentanza più ampia nel parlamento. Nell’elettorato del Movimento esistevano però componenti con profili sociali e atteggiamenti molto differenziati. La formazione del primo governo Conte con il sostegno della Lega aveva creato molti problemi al M5s. Il movimento incontrava notevoli difficoltà a realizzare i punti più importanti del suo programma e a soddisfare le attese dei suoi elettori. La Lega invece aumentava in misura rilevante i suoi consensi perché combinava i tradizionali programmi del centrodestra con la rappresentanza della protesta popolare. La formazione del secondo governo Conte con il sostegno del Pd e dei partiti di centrosinistra non permetteva al M5s un recupero significativo dei consensi elettorali perduti. Le possibilità di avviare nuove politiche erano d’altra parte molto limitate per gli effetti della pandemia. Il secondo governo Conte fu costretto a dimettersi per il passaggio all’opposizione di Matteo Renzi che temeva un consolidamento della alleanza del Pd con il M5s. Il sostegno al governo Draghi, creato per fronteggiare con più efficacia la pandemia, ha creato nuove e più significative difficoltà al movimento. Se la partecipazione ai due governi guidati da Conte aveva messo in evidenza le difficoltà di realizzazione dei propri programmi, nel governo Draghi il ruolo del M5s appare sempre più marginale. Il suo elettorato si è ridotto in modo significativo, con la perdita delle opzioni di voto di molti elettori che si attendevano cambiamenti radicali della politica. Dunque, dal 33% delle politiche 2018, si è passati prima al 17% delle Europee 2019 (quasi dimezzati) e poi alle attuali stime sulle intenzioni di voto, che vedono il M5s intorno al 13-14%. In pratica, in quattro anni, il Movimento ha perduto per strada quasi il 60% del proprio elettorato. Così come era stata davvero impressionante l’ascesa, altrettanto impressionante è stata la crisi di consenso. E dove sono finiti quegli elettori che avevano così fortemente creduto nella loro proposta e nel loro ruolo inedito nello scenario partitico? Soltanto un terzo circa gli è rimasto fedele. Si tratta di uno dei tassi di fedeltà più basso mai registratosi nel breve volgere, appunto, di quattro anni. La maggior parte dei fuoriusciti (oltre il 22%) è andata verso destra, in particolare verso i Fratelli d’Italia, transitando provvisoriamente dalla Lega in occasione delle Europee e approdando poi definitivamente nel partito di Giorgia Meloni. Pochi, molto pochi hanno scelto il Partito Democratico (poco più del 4%) o altre forze di centro-sinistra o di sinistra, con cui peraltro si è aperto un dialogo rilevante quanto difficoltoso per impostare un percorso futuro comune. La maggioranza relativa del 40% è (tornata) verso l’astensionismo o l’indecisione, in qualche modo confermando la profezia originaria di Beppe Grillo, che intendeva dare una voce a coloro che, insoddisfatti della politica della “casta”, si sarebbero tenuti lontani dalle urne. E che ora probabilmente torneranno di nuovo verso l’astensione, delusi dalle loro scelte di cinque anni orsono. Talmente delusi che, oggi, il loro giudizio sul Movimento 5 stelle appare nettamente negativo: solo il 15% tra loro, al pari di chi è andato verso destra, ne dà infatti una valutazione positiva. Non si salva nemmeno il nuovo leader del M5s: Giuseppe Conte, molto amato quando era Presidente del Consiglio, viene ora giudicato positivamente solo da poco meno del 30% dei protagonisti della diaspora. In questo paper si cercherà di comprendere più in profondità da una parte le scelte compiute dagli elettori del Movimento 5 stelle dal 2018, passando attraverso gli anni della pandemia e dei due governi Conte, fino ad oggi e, dall’altra, le caratteristiche strutturali delle diverse anime e le motivazioni dei mutamenti che hanno visto il progressivo deterioramento della fiducia verso una forza politica che, fino a poco tempo fa, aveva rappresentato un possibile punto di svolta nel panorama politico italiano. Keywords: Movimento 5 stelle; governi Conte; flussi di voto

Concerns about the consequences of social inequality have been on the rise over the past years. Negative outcomes of inequality have been described in many realms of social and political life, from population health to social trust and social capital, and from democratic attitudes to levels of civic and political participation. Increasing social inequality is often credited in many public debates to be the cause of electoral earthquakes, such as that occurring in the 2018 Italian general election. Less frequent are studies on the effects of inequality on vote choice. This paper will address this issue from three angles. First, it will examine the ecological correlation between income inequality variation (Gini index) at the level of Italian provinces and the electoral results of the main parties at the same level. Second, it will explore the relationship between the propensity to vote for the same parties at the individual level and provincial income inequality variation. Third, it will model how varying levels of income inequality condition the effects of microlevel determinants in discriminating the vote choice across all parties (global effects). This entails a so-called generic approach to party choice through multi-level statistical modelling of a stacked data matrix. This analytical and methodological approach allows us to assess not only the direct effect of income inequality on vote choice, but also the conditioning effect of Gini on the relative weight different individual characteristics exerts on vote choice. Hence, the focus shifts from the explanation of party-specific vote choices to a systemic explanation of the voting decision process (the voting calculus) when contextual characteristics such as income inequality are taken into account. Income inequality is estimated from the aggregated income brackets of the 2017 tax declaration provided by the Italian Agenzia delle Entrate. Electoral data at the ecological level come from the Ministry of Interior archive and those at the individual level from the Italian study of the 2018 election which is part of the Comparative Study of Electoral Systems (CSES). Results indicate that income inequality moderates the global effects of some microlevel predictors on vote choice, but less than what could be hypothesized on the basis of the effect of Gini on party-specific support measured both at the individual and especially at the ecological level. In particular, income inequality diminishes the importance of the traditional political anchors of vote choice, with potentially negative consequences for the quality of the voting decision process.

Under what conditions in times of exceptional global crisis do citizens find restrictive and punitive measures from the state as legitimate? Previous studies have largely assessed citizens’ responses to crises such as pandemics and changes in their political attitudes during enhanced threats such as terrorism. However, scholars have rarely employed experimental approaches that rely on a comparative perspective between different types of crises and of limitations of individual liberties in order to estimate the relative weight of various factors in explaining citizens’ political postures. To this end, we administered a conjoint survey experiment to a nationally representative sample of 1.000 Italian citizens. We presented respondents with a vignette where we randomly vary three types of crises (terrorism, climate change, pandemic) along with its severity, and the measures proposed by states to contrast the threat (punitive or restrictive measures, enhanced controls). Our results show that Italian citizens find limitations to their individual liberties more legitimate when the crisis is more severe, and when such measures punish who does not respect rules. Moreover, and supporting what recently literature has underlined (Schmidt, 2021), our results highlights public reactions similar when comparing climate change and pandemic crises, contrary to what happens with terrorism. Working on various interactions between the factors at play in the experimental vignette and on citizens’ sub-group characteristics (such as ideology, populism, and age), we show how some important factors drive such results in different directions. Our study generates interesting implications for the study of citizens’ attitudes in terms of risk perception, political crises, and individual freedoms. Moreover, our analysis also contributes to policy-discussions about the legitimacy of restrictive, punitive measures, and effective ways to frame restrictions.

Fra gli effetti della globalizzazione che si sono maggiormente affermati, in particolare presso le fasce sociali più deboli delle società liberal-democratiche occidentali, i cosiddetti “perdenti della globalizzazione” (Kriesi, 1998 e 2006), si registra da qualche tempo una sorta di avversione sistematica verso tutto ciò che ha a che vedere con le scelte maggioritarie di cittadini e opinione pubblica, che si sostiene vengano di fatto veicolate dai “poteri forti” e dall’establishment politico-economico. Vi è un’ampia letteratura sociologica e politologica che sottolinea come l’affermazione dei processi della globalizzazione abbia favorito una presenza sempre più pervasiva di ruoli tecnici nei luoghi della decisione politica. Tecnici e Agenzie di tecnici che, in determinate situazioni, specie in condizioni di crisi ed emergenza, si sostituiscono agli abituali organi istituzionali preposti alla decisione politica, generando implicitamente una condizione di “deficit democratico” (Moravcsik, 2004; Decker, 2002; Norris, 2011; Rumford, 2003; Merkel et al. 2003; Merkel 2014). È in questo modo che la politica acquisisce una veste sempre meno influenzata dal voto popolare, e così facendo si distacca dal popolo degli elettori e dalle sue aspettative (Mounk, 2018). Un aspetto al quale, nell’emergenza sanitaria procurata dalla diffusione del Covid19, si è aggiunta una maggiore propensione, a livello governativo, ossia dove il potere politico si è maggiormente concentrato negli ultimi due anni in gran parte delle polity democratiche, a prescindere dalla specifica forma di governo, a prendere decisioni restrittive per tutelare la salute pubblica, con il supporto di organi tecnici che hanno giustificato questi interventi in ragione della necessità di contenere il diffondersi del contagio (Musella, 2015). E ora, con la guerra condotta dalla Russia in Ucraina, questo latente disagio presente in un’area significativa della popolazione rischia di sommarsi a una nuova situazione di emergenza, con evidenti conseguenze soprattutto su aspetti della vita quotidiana (inflazione, costo della vita, prezzo dei carburanti e dei generi alimentari ecc.), che potrebbe originare una crisi sociale ed economica in grado di avanzare una sfida ancor più impegnativa per le nostre democrazie. La verticalizzazione del processo decisionale prodotta dall’emergenza sanitaria ha favorito la formazione di movimenti “anti-sistema” che hanno fatto del rifiuto della vaccinazione e della contestazione del GreenPass la loro bandiera (Citro, 2021). Si tratta peraltro di un fenomeno che viene da lontano e che trovava prime forme di manifestazione già negli anni precedenti la pandemia, come dimostra il caso italiano dei Forconi, guidati dallo stesso Pappalardo protagonista oggi della contestazione al GreenPass, e quello francese dei “Gilet Gialli”, movimento nato per protestare contro l’aumento dei prezzi dei carburanti e l’assenza di misure sociali di sostegno ai soggetti più colpiti dalla crisi economica (AA.VV., 2019). Le manifestazioni dei No Green Pass, presenti in molte città italiane (ed estere) per molte settimane, hanno coinvolto solo qualche migliaio di persone. Ma le loro proteste hanno alimentato un latente scetticismo in diversi settori dell’opinione pubblica, oltre che sui social media, in molte testate giornalistiche e trasmissioni televisive, che hanno contribuito a diffondere dubbi e incertezze rispetto all’efficacia delle campagne di vaccinazione e di altre misure di prevenzione, quale il Green Pass. E se dunque non sono (stati) molti coloro che si sono mobilitati “fisicamente”, per i costi della mobilitazione, dietro quei pochi attivisti vi sono milioni di italiani che, in qualche modo, la pensano più o meno come loro, considerando il Green Pass, così come le altre misure di prevenzione del contagio, “misure esagerate, che violano la libertà di chi non vuole farsi vaccinare e mirano a restringere gli spazi di agibilità dei cittadini, prefigurando una forma strisciante di dittatura”. Diventa perciò interessante cercare di capire chi sono, quali sono le loro caratteristiche in termini demografici e politici, e in che misura questo orientamento si correla ad altri atteggiamenti di critica incondizionata verso la società e le sue regole di convivenza. Utilizzando un campione sufficientemente numeroso di elettori (circa 4mila individui interrogati negli ultimi due mesi da Ipsos) cercheremo di tracciare un quadro significativo del loro profilo, confrontandolo poi con chi non si dichiara contrario al Green Pass, al fine di pervenire alla costruzione di una tipologia, differenziando i “No Green Pass” non vaccinati da quelli vaccinati. Cercheremo inoltre di analizzare l’andamento tendenziale di questi atteggiamenti nel corso dell’ultimo anno e di correlarli con le principali caratteristiche strutturali ed il corrispondente orientamento politico. Da ultimo proveremo ad indagare l’esistenza di un’ulteriore correlazione con le valutazioni fornite rispetto alla guerra in corso in Ucraina, con riferimento al riconoscimento delle ragioni del popolo ucraino o della Russia di Putin. Scopo principale del nostro lavoro, più in generale, sarà quello di tracciare una sorta di “fenomenologia dei contrari per scelta”, al fine di evidenziare come questi “individui in rivolta” possano essere la punta dell’iceberg di un’ondata di crescente profondo dissenso, con la quale occorrerà confrontarsi seriamente, per evitare che quell’iceberg emerga in maniera ancora più evidente nel mare della nostra società, mettendo in discussione alcuni premesse fondamentali della nostra forma di convivenza democratica. E come questo fenomeno, anche a seguito della guerra intrapresa dalla Russia in Ucraina, cioè nel momento in cui si profila un confronto fra autocrazie e democrazie a livello globale, possa rappresentare una seria criticità per il supporto alle democrazie occidentali e alla loro cultura politica. Keywords: No GreenPass; Movimenti di protesta; Tecnocrazia

Institutional confidence, also known as institutional trust, is a form of political support (Easton 1965, 1975) encompassing evaluations about regime legitimacy, institutional and authorities’ performance, and other expectations and evaluations. Despite conflicting conceptual and theoretical argument about its nature (for a review, see Schnaudt 2019), individual confidence in political institutions is conceived here as an inherently mixed form of political support (Torcal and Montero 2006), whose more diffuse or specific nature is conditional on individual characteristics and contextual factors. This paper aims, then, to contribute to this debate by empirically investigating the societal and (intra)individual dynamics of institutional confidence in Italy during the Covid-19 pandemic. Existing descriptive studies (Ladini 2021) have already shown that during this historical period Italians’ confidence in institutions has been rather volatile, following patterns similar to those observed in other European countries (Bol et al. 2021; Schraff 2021; Esaiasson et al. 2021; Jennings 2020; Kritzinger et al. 2021). To what extent these dynamics can be attributed to the varying importance of partisan stances for citizens’ evaluations of and expectations about Italian political institutions? Can the varying impact of these stances be attributed to contextual (societal and/or political) exogenous factors? By relying on the 2020-2021 ResPOnsE COVID-19 cross-sectional and panel survey data, this paper seeks to answer these research questions, thus contributing to the broader debate about the nature of individual confidence in political institutions.

Much work on party strategy assumes that voters will reward particular issue stances with their ballot. Vice versa, much work on voting behavior assumes that parties’ issue stances are designed with vote maximization in mind. We integrate these assumptions into one dynamic model: Parties first strive to emphasize certain policy issues according to their potential to attract new voters while defending existing support—the strategic sweet spot described by “Issue Yield” theory. Among voters, sensitive to such issue priming, these issues acquire higher salience as criteria orienting the voting decision. As a result, the electoral return on effective Issue Yield strategy will manifest itself in a competitive tilt toward the winning party. Overall, we expect vote switching to occur when there is a disparity in the degree to which parties pursue their Yield. We test these hypotheses using data from the Issue Competition Comparative Project, which combines seven pre/post-election voter surveys from Europe and the US - running across the COVID crisis - with systematic codings of parties’ Twitter campaigns

Under what conditions do marginalized ethnic minorities vote based on economic inequality concerns over their ethno-national identity? In an inequality-based ethnic vote equilibrium, the material economic promises made to minority voters increase the opportunity cost of voting based on ethnic identity. Because coalition-targeted minorities stand a legitimate chance at having their interests represented in government, voters change their strategic calculus and support parties promising to reduce economic inequality. Using a mixed methods approach, I show that during Israel’s first post-COVID legislative election in 2021, an increase in the supply of intra-systemic economic redistribution platforms and decrease in the salience of Israeli-Palestinian Conflict concerns was met with widespread Arab voter demand for greater intra-systemic economic inequality mitigation measures, resulting in the first Arab policymakers entering Israel’s ruling coalition and securing increased development funding for Arab communities. This research has important implications for reducing inequality among ethnic minorities and increasing their legislative representation more broadly. Key words: political economy, inequality, redistribution, Israel.

After a rapid proliferation in the United States over the last decade, research on affective polarization has finally started to also take off in the rest of the world. System level studies have demonstrated that affective polarization is undoubtedly present in multiparty systems and the intensity of partisan loathing in the US context is not exceptionally high in comparative perspective. In many countries, a significant share of highly negative partisan feelings is associated with populist radical right parties that have a very strong polarizing appeal. However, there is a striking pattern that has hitherto remained in the background in affective polarization studies: the hostility between the populist right and the other parties is clearly asymmetric, i.e. the right-populists receive much more dislike from the supporters of other parties than vice versa. Moreover, this intense loathing towards the right-populists goes beyond what ideological positions would predict. Even the voters of the parties that do not stand so far from the populist right on the left-right scale or do agree with them on cultural dimension issues such as (curbing) immigration, tend to exhibit highly negative feelings toward these parties and their voters. Thus, in the coattail of the rise of right-wing populism across the world, a potentially even stronger anti-populist sentiment has also emerged. While some authors have noticed this pattern, it has remained as a minor observation in most of their work. Therefore, while the phenomenon has been detected in a wide range of countries, we do not know how much it varies cross-nationally and what could be the factors that drive this asymmetry in feelings between the right-populists and the rest of the partisan field. In this paper, we aim at addressing this gap in affective polarization research, focusing on the European countries to ensure better data availability and comparability of cases. Using data from the Comparative Study of Electoral Systems (CSES), we map the levels of asymmetry in affective polarization towards the populist right parties across the European countries. Having determined how uniform this phenomenon actually is and how much it varies cross-nationally, we then move the analysis to the individual level to find out which variables best predict the feelings towards the right-populist parties and their supporters. Combining data from CSES Wave 5 and our own survey that is going to be launched in France, Italy, Hungary and Estonia between February and June 2022, we test the effects of a range of attitudes regarding populism as a (thin) ideology, populism as a political style, nativism, cultural authoritarianism, illiberal democracy and Covid-19. Letting respondents place parties on different political dimensions, we also scrutinize the possibility that other voters perceive right-populist parties to be more extreme than their own supporters do. This study will contribute to a better understanding of the place that the populist parties occupy in the partisan affective landscape, and help us better grasp which conditions are conducive to higher accommodation (or rejection) of the right-populist parties by other political actors.

The concept of affective polarisation has a central role in American Politics but there is evidence suggesting that affective polarisation is prominent in many European countries as well (Wagner 2021). In the American context, studies have suggested that women are more affectively polarised than men and this is due to gender differences in issue positions but also due to the impact of gender on shaping political identities (Ondercin,2020). In this paper, I focus on the case of Europe, and by employing data of more than a million citizens living in Europe from the late 1980s until 2020 I examine the impact of gender on affective polarisation outside of the American context.

Do affectively polarised people vote or stay home on election day? Although there is an increasing number of comparative studies focusing on the origins of partisan affective polarisation, our knowledge about its impact on individuals’ decision to vote is still limited. This article takes a closer look at the relationship between propensity to vote and partisan affective polarisation by distinguishing those produced by in-party like and by out-party dislike. The results show that both in-group like and out-group hate sentiments increase people’s propensity to vote, and that the effects of the latter are more accentuated. The argument presented in the following pages is based on the analysis of a two-wave panel study conducted in Spain between April and May 2019.

A growing body of evidence shows that affective polarization is on the rise in the US. However, comparative longitudinal assessments found little evidence of a generalized increase in affective polarization in multi-party systems. Instead, the mixed results highlight the inconsistencies in trends across countries, disconfirming the existence of a pattern of increased affective polarization over the last decades across Western democracies. This study adds to the present literature in three ways. First of all, we expand the number of countries and the time span of the analysis, in view of providing the most encompassing comparative and longitudinal account of affective polarization. To do so, we resort to a newly assembled dataset able to track partisan affect in 18 Western democracies over the period 1961-2020. Secondly, we provide a disentangled assessment of the two components of affective polarization, namely, in-group affect and out-group contempt, in order to understand which specific component (if any) is responsible for changing polarization patterns over time. Thirdly, we bring forward the consideration of the direction of affective polarization, and show that the major change over time can be attributed to its increasing negativity.

In the United States, a lively scholarly debate on the nature of mass polarization continues. After much ink had been spilt on the issue of ideological polarization, scholars of American politics have recently turned their attention to worrying trends in affective polarization – the tendency of Democrats and Republicans in the mass public to “dislike, even loathe, their opponents” (Iyengar, Sood, and Lelkes 2012, 405). The potential causes of this growing partisan animosity are many and still hotly debated, but one interesting suspect is the phenomenon of “social sorting.” Over the last several decades, the two major parties in the United States have become increasingly homogenous in terms of their social composition. Ethnic minorities, residents of urban areas, and non-religious individuals have sided with the Democrats; white, religious men living in rural areas have associated with the Republicans. These are, of course, crude generalizations – but precisely the type of crude generalizations that social identity theory predicts people will make when forming mental images of in-groups and out-groups (see Ahler and Sood 2018). This process of social sorting “has made it much easier for partisans to make generalized inferences about the opposing side” (Iyengar et al. 2019, 134). From the perspective of a socially sorted partisan, the opposing side is emphatically not amorphous: it is composed of precisely the sort of people with whom one does not associate. The result has been the formation of a partisan “mega-identity”, “with each party representing not only policy positions but also an increasing list of other social cleavages” (Liliana Mason 2018, 20). Cross-cutting identities, long thought to be a stabilizing force for democratic regimes, consequently wither away as a social chasm opens up. In Canada, there has been only “sporadic interest” in studying mass polarization (Merkley 2021a, 269; see also Gidron, Adams, and Horne 2019). In particular, the social dimension of polarization has been almost entirely ignored. Merkley (2021b) represents the sole exception. Using cross-sectional data from 2019, he finds relatively limited levels of actual social polarization, but substantial levels of what he terms “false social polarization.” Just as in the United States, Canadians perceive political parties to be much more homogenous than they actually are and vastly overestimate the share of a party’s supporters who are members of party-stereotypic social groups. In other words, partisanship is not, in fact, neatly organized with other social identities, but Canadians perceive it to be. What is still unknown is whether actual social sorting has increased over time. That is, have Canadians’ partisan identities become increasingly aligned with their other social identities? This project aims to help fill this gap by studying the history of social sorting along partisan lines in Canadian federal politics, as well as the link between social sorting and affective polarization. The American case will also be studied in order to act as a sort of "baseline" against which to compare the Canadian situation. In potential extensions (which I may or may not have time to prepare in time for September), I intend to explore a broader cross-section of advanced democracies by leveraging CSES data. Should we expect the patterns found in the American case to travel north? After all, the Canadian party system has long been presented as a “deviant case” (Johnston 2017, 3). The very concept of party identification has often left Canadian political scientists dubious (see e.g. LeDuc et al. 1984). Yet many of the developments observed in the United States appear to be reflected in the Canadian context. Longitudinal, cross-national data has revealed substantial levels of affective polarization across different party systems in advanced democracies, with Canada exhibiting a noticeable upward trajectory in recent decades (Boxell, Gentzkow, and Shapiro 2020; Johnston 2019; Reiljan 2020). The question that remains is why. As much of the literature on mass polarization focuses on the United States, most explanations that have been advanced are US-centric. As a result, many of these potential causes of affective polarization do not travel well beyond the American context. For instance, the role played by the partisan news media (see e.g. Lelkes, Sood, and Iyengar 2017; Levendusky 2013) most likely does not apply in Canada, where mainstream, non-partisan news outlets dominate. By contrast, the theory of social sorting presented above relies on mechanisms that are “plausibly universal” (Harteveld 2021). Are those mechanisms likely to be at play in Canada? The structure of the Canadian party system is fundamentally different than what is found in the United States and, indeed, in most advanced democracies. The two main parties in Canada – the Liberal party and the Conservative party – have dominated electoral politics for 150 years by adopting the approach of “brokerage politics”: they seek to represent a national constituency rather than cater to particularistic interests. As such, the Liberals and the Conservatives represent “one of the great exceptions” to the Lipset-Rokkan paradigm, which centers discussions of party systems around the relevant cleavages that divide the parties (Carty and Cross 2010, 193). From the perspective of social sorting theory, therefore, we would expect the major parties to be composed of large, socially heterogenous coalitions that make for numerous cross-cutting cleavages. To be sure, electoral cleavages did exist, most notably those divisions based on religion and language; still, most accounts of the Canadian party system stress the absence of cleavages and the broad base of support enjoyed by the major parties. But this neat picture is muddled considerably when one considers the “electoral earthquake” of 1993 (Johnston 2008, 816) and the multiparty competition it spawned. From that point onward, a major party in Ottawa now competed exclusively in a single province, while a succession of insurgent conservative movements catered to Westerners. This fragmentation of the party system, absent for most of the 20th century, may have set the stage for social sorting of the sort described earlier. In order to examine this phenomenon, this project will employ a two-pronged approach. First, from a descriptive standpoint, I will examine the extent to which partisan groups in Canada have become more or less socially sorted over time. The approach here will be to use historical data from the Canadian Election Study (CES) in order to build individual-level estimates of cross-pressures that emanate from different social identities, based on work by Brader, Tucker, and Therriault (2014). The procedure is to estimate regression models that relate demographic characteristics – language, region, social class, ethnicity… – to partisan preferences; to compute predicted probabilities of supporting different parties for each survey respondent; and finally to use the variance in predicted probabilities for a given respondent in order to estimate the intensity of cross-pressures that emanate from the social group memberships that were used as input to the regression models. The rough idea is that a respondent who is assigned, based on the statistical model outlined above, a 0.55 probability of voting for the Conservatives and a 0.45 probability of voting for the Liberals faces greater conflicting pressures on the basis of his or her social identities than another respondent who is assigned 0.75 probability of voting for the Conservatives and a 0.25 probability of voting for the Liberals. If Canadians have socially sorted in a manner similar to the United States, the latter type of voter should be more frequent than the former. By leveraging CES data going back to the 1980s, I will be able to trace the history of cross-pressures emanating from social group memberships and determine whether or not the pattern of increasing homogeneity found in the United States is also present in Canada. In terms of empirical presentation, visualizing the distribution of cross-pressure scores for each iteration of the CES should be sufficient. For a more formal test, I will also conduct Kolmogorov-Smirnov tests, which examine the null hypothesis of identical cumulative distribution functions between two distributions. Finally, given the pattern of asymmetric social sorting in the United States, with Republicans growing homogenous at a faster rate than Democrats (Mason 2018), presenting the distributions conditional on party identification could also be fruitful. The second part of the project will directly examine the key relationship that makes social sorting a relevant object of study: the idea that highly sorted partisans are less likely to be able to engage socially with their political opponents and therefore more likely to be affectively polarized. I will retain the cross-pressure scores introduced above as a key variable and introduce it on the right-hand side of a regression model that seeks to predict affective polarization, measured as the difference in feeling thermometer scores between a respondent’s in-party and out-parties. Here, I am still undecided between two empirical strategies. The first is to run cross-sectional models while attempting to control for all relevant confounders. The second is to exploit the panel maintained by the CES from 2004 to 2008, which spans 3 election cycles. While this would allow for more credible causal claims, I am undecided as to whether it makes sense to treat within-individual, across-years differences in cross-pressures as “real”, given that some of the identities under examination are essentially “fixed” (e.g. gender).

One of the defining features of liberal democracy in the 2000s was the ability of the political parties to reach across the political divides and the willingness to talk about it and show mutual respect. In the past decade, however, this practice has seen a significant decline across much of the liberal world. Unlike liberal democracies, hybrid regimes have been marked by asymmetry between ruling and opposition parties. Nonetheless, similar to liberal democracies, hybrid regimes such as Hungary and Turkey are also becoming increasingly more polarized. Furthermore, the lack of effective balancing political institutions intensifies the costs of polarization for opposition groups. As a result, in many hybrid regimes, we see the increasing dominance of some political parties and the inability of the opposition to mount a substantial challenge. Furthermore, the incumbent's rhetoric further fuels this divide, pushing apart the different social groups. The hybrid regime literature has also taught us that effective pre-electoral coalition formation among opposition parties can threaten the incumbent. The literature has examined how international actors or the diffusion of democracy affected these processes. What is lacking is the cause that would spur the opposition elites to organize themselves. In this paper, we argue that political polarization could increase the likelihood of coalition formation among opposition parties in hybrid regimes. Building upon descriptive evidence around the world, we expect that in a hybrid regime environment where the incumbent substantially controls the media, polarization will decrease the visibility and resources of opposition actors. In turn, such change in power dynamics will give social legitimacy and incentives to opposition parties to form electoral coalitions. We fit a fixed-effects logistic regression model to test the paper's hypothesis, using all hybrid regime-years after the cold war as a sample. Our dependent variable is the opposition coalition. Our data on opposition coalitions is an updated version of Donno's (2013) dataset on opposition coalitions in electoral authoritarian regimes. Our independent variable is the political polarization index from the Varieties of Democracy Institute data. We also include other variables that explain coalition formation, such as the electoral system, whether the political system is presidential, and the government's economic performance. The results suggest that there is a positive and statistically significant relationship between the degree of political polarization and the likelihood of coalition formation. We aim to make two contributions. First, we show that polarization need not be harmful for certain democratic qualities in hybrid regimes. Instead, it can serve as a stimulus for opposition groups (parties or movements) to unify. Consequently, and second, we aim to show that incumbent-driven political polarization can contribute to increased coordination among opposition actors. These findings could counterintuitively imply that political polarization can be an opportunity for democracy promoters. In turn, this can contribute to the democratization of hybrid regimes by-elections. Key words: polarization, hybrid regimes, opposition coalitions, parties, democratization

The last two decades have seen a decline in the overall support for the European Union (EU) and a concomitant rise in Eurosceptic parties in member states. While both have been partially explained in terms of economic downturn, here we focus on the extent that national economic contexts shape citizens of Europe to vote for parties that hold clear anti-EU views. Given the literature, and particularly following the 2007/8 financial turned economic recession, the expectation is that negative economic conditions produce declining trust in the EU and thus directly shape individuals’ anti-EU vote choice. Yet, the extant literature lacks a consensus. Therefore, using the European Social Surveys, we examine individual-level data on vote choices at the national-level merged this with a wide array of macro-economic variables over 2006-2018 to determine the extent to which economic contexts shape Eurosceptic voting. However, controlling for both individual- and national-level variables, while national-level economic variables are implicated in individuals’ Eurosceptic vote choices, the evidence is far from unequivocal, and in the context of specific macro-economic variables, contrary to our theoretical expectations.