Personality Goes a Long Way

Abstract

This article reviews books which test the personalization of politics, looking at different dimensions of the growing importance of leaders over time, namely for political parties, in electoral behaviour and in the media. Only recently have wide-ranging comparative longitudinal studies on leaders been carried out. The personalization thesis is not equally demonstrated across all dimensions. Indeed, we find something of a puzzle: There is no strong trend towards personalization of party organizations, whereas in electoral behaviour the evidence points to the increasing use by voters of leaders as heuristics. This attests to the decline of the importance of parties. The personalization of media may be the mechanism which explains the change in voting behaviour, and the third and final section of the review looks into that arena. We conclude with some suggestions on further research on the personalization of politics.

William Cross and Jean-Benoit Pilet (eds), The Politics of Party Leadership (Oxford: Oxford University Press, 2015).

Alessandro D’Arma, Media and Politics in Contemporary Italy (London: Lexington Books, 2015).

Diego Garzia, Personalization of Politics and Electoral Change (London: Palgrave Macmillan, 2014).

Ana Inés Langer, The Personalisation of Politics in the UK: Mediated Leadership from Attlee to Cameron (Manchester: Manchester University Press, 2011).

Gianluca Passarelli (ed.), The Presidentialization of Political Parties: Organizations, Institutions and Leaders (London: Palgrave Macmillan, 2015).

In January 2017 Donald Trump was inaugurated as the forty-fifth president of the United States. Trump’s ability to win the Republican Party nomination, against the will of the party grandees, went against the received political science wisdom, which placed party elites in charge of the choice of presidential candidates in the US (Cohen et al. 2009). His subsequent victory against Democratic candidate Hillary Clinton further challenged the idea that the two parties controlled access to American institutions. Trumpism is a clear sign of the decline of political parties as institutional gatekeepers and is symptomatic of the rise of the media-driven, outsider leader. Yet is this a specifically American phenomenon, or has it spread to other countries?

In Italy, the rise of Silvio Berlusconi as leader of Forza Italia and the longest-serving prime minister of Italy is perhaps the closest parallel to Trump. Berlusconi was also an outsider – a media and construction billionaire – who stormed Italian politics. He was elected as MP in 1994 and went on to serve on three different occasions as prime minister of Italy (1994–5, 2001–6 and 2009–11). Berlusconi, unlike Trump, created his own party, at a time when the party system in Italy was imploding (Bartolini et al. 2004) under the weight of tangentopoli. Despite being dogged by the judiciary for most of his mandates, he dominated politics in Italy for more than a decade.

Another Italian politician, Beppe Grillo, the leader of the Movimento 5 Stelle (5 Star Movement – M5S) is also a clear example of a mediatized personality, in this case using the internet to gain visibility. His party has been described as belonging to a personal party model (Diamanti 2014). He illustrates the importance that new media may have in the process of the personalization of politics which has been recurrent in Italy. Founded in 2009, M5S won 25 per cent of the vote in the 2013 legislative elections and 109 seats in the Italian parliament, becoming the second largest party in Italy.

The election of Marcelo Rebelo de Sousa as president of Portugal can also be counted as one of the more recent cases of the extreme mediatization of politics. Despite not being the head of government, the president of Portugal holds important prerogatives both in terms of veto power and in relation to the dissolution of the Assembly, which can be crucial when governments are weak (Amorim Neto and Lobo 2009). Although Rebelo de Sousa has been a centre-right party member for most of his active life, and was even leader of the Partido Social Democrata (PSD) between 1996 and 1999, Marcelo – as he is known to the Portuguese – became a household name from 2000 because of his weekly political commentary shown on open access television networks. Marcelo is, to a large extent, a product of the media, where he carefully crafted an image of a likable politician over the course of 15 years. In 2015 he decided to run for the presidency against the wishes of the PSD leader, Pedro Passos Coelho, who saw him as an outsider. He persevered nonetheless, running a campaign with little funding and winning the presidential election on the first round in January 2016 with 52 per cent of the vote.

Beyond the process of mediatization, another common phenomenon is the growing autonomy of leaders from the party organization across the ideological spectrum. Indeed, the rise of Jeremy Corbyn as leader of the British Labour Party may be signalling the same kind of decline of party structures, albeit at the opposite end of the party spectrum. A long-standing MP and representative of the most left-wing group in the Labour Party, Corbyn’s election in the summer of 2015 was partly a story of the party grassroots rejecting the candidates with mainstream party support. Just as the Republican Party was surprised by Trump, the venerable 115-year-old Labour Party did not really think Corbyn could win (Wintour and Watt 2015).

The trend of new leaders who criticize the mainstream political elites can also be found elsewhere. We see it in the case of Beppe Grillo described above, but also more recently in Southern Europe with the rise of Alexis Tsipras in Greece (Baboulias 2016) and Pablo Iglesias, Podemos leader in Spain. They are examples of leaders who dominate their parties and represent a left-wing drive to reject the political status quo.

These examples show that the emerging trend of ‘outsider leaders’ should not be exclusively associated with the right wing of the party spectrum. Nor should it be confused with the phenomenon of populism, which includes other characteristics (Mudde and Rovira Kaltwasser 2012). The rise of outsider party leaders may be seen partly as a corollary of the decline of political parties which has been occurring in the last few decades in advanced industrial democracies. This decline has been measured in terms of membership (van Biezen et al. 2012), but can also be seen in the decline of trust in political institutions (Dalton 2013). It is no coincidence that Berlusconi emerged in Italy in the mid-1990s. Indeed, at that time, the Italian party system was imploding due to a large-scale judicial investigation into corruption which affected all parties who had been involved in government since the end of the Second World War.

More generally, the trend towards an increasing distance between political parties and society has paved the way for a looser relationship between citizens and politics, making the latter more open to political entrepreneurs. Parties have evolved from society-based actors, with strong links to civil societies in advanced democracies in the 1960s and 1970s, to catch-all formats that were more adapted to the growing individualization of society (Krouwel 2006). This ‘divorce by mutual consent’ evolved in the last few decades into a cartel system of state-dependent political parties (Katz and Mair 1995). Arguably, the progressive hollowing-out of political parties as territorial organizations has contributed to the rise in the importance of the media as intermediaries for establishing relations between political incumbents and society. It has also favoured leader-centric developments in election campaigns as well as government organizations (Poguntke and Webb 2005).

In addition, it is important to note that in several countries where the phenomenon of personalization of politics was already apparent, namely Italy, Portugal, Spain, Greece or the UK, the last few years have been characterized by economic crisis and continued recession, due to the onset of the Great Recession in 2008. Indeed, the financial crisis that began in the US in 2008 and quickly became a crisis of public finances in Europe meant that countries with larger public debts and underlying weaknesses in their economic indicators found it increasingly difficult to access the financial markets. In Europe, politicians in most countries had to impose austerity policies to fulfil external commitments to an increasingly dissatisfied electorate. The crisis which took place in the eurozone from 2008 onwards may have compounded feelings that existing mainstream parties simply cannot deliver the prosperity which citizens have become used to, in turn rendering citizens more willing to listen to outsiders who can now criticize the mainstream not only for their cartel behaviour in organizational terms, but also for their lack of positive economic results (Lewis-Beck and Lobo 2017).

It is, however, important to understand that this phenomenon of leader-centric politics is not confined to the populist or new parties, nor should it be. Across Europe, for the last decades we have witnessed the appearance of several media-centric leaders, such as Tony Blair in Britain, Nicolas Sarkozy in France, or José-Luis Rodríguez Zapatero in Spain, and there have been important studies on this topic (Aarts et al. 2013; Poguntke and Webb 2005). Indeed, there have been attempts to track the personalization of politics from an organizational, campaign and voter perspective. More recent developments concerning the appearance of outsider leaders highlight the extent to which advanced democracies and their electorates are open to such formats. Indeed, given the recent developments in Europe and elsewhere, the fundamental question posed in this review article is: How can we understand the leaders’ place and their growing importance in democratic political systems?

The books we are reviewing are important not only given the recent political developments, but also given the fact that mainstream political science has tended to neglect the study of leadership. Indeed, there has not been nearly enough academic work on this topic. The absence of work arises from a combination of factors. Leaders and leadership are concepts that are close to the concept of power and are inherently difficult to quantify and measure. In the last decades, the quantification drive in political science has, thus, made it harder for leaders to be at the centre of political analysis. Instead they have often been neglected in favour of more easily quantifiable aspects of political science. Not only that, but the fact that much of what is understood as political science today grew out of political sociology has meant that a sociological perspective on politics has informed much of the discipline, especially when we discuss its electoral dimensions as well as the party literature. Still on the methodological front, the progressive fragmentation of different aspects of political science hampers the understanding of the role of leaders, as those who focus on public policy seldom dialogue with electoral studies specialists or party researchers. Whereas leaders may have an impact on all these arenas, there is a lack of integrated studies aiming to understand leaders.

That said, the problem of leadership studies is not derived solely from methodological reasons. Whereas in other areas of study where leaders matter, such as business studies, the concept of leadership has remained very firmly at the centre of research, in political science it has been residual. This is partly due to normative ideas about democracy that formed in the light of European political experiences at the beginning of the twentieth century and which have fundamentally shaped the way politics is conceived in the discipline today. Following the leader-centric totalitarian and authoritarian regimes which led to two world wars in Europe, there was an urge to promote democracy as a process – vide David Easton’s conceptions of the political system. Since the 1980s, attempts to place institutions at the centre of political life moved the discipline away from sociological interpretations of politics and rarely emphasized the role of leaders within institutions.

Several recent review articles on leaders and personalization have noted a number of these key problems (Barisione 2009; Garzia 2014; Lobo 2014). Even so, some scholars have researched the growing importance of leaders – at elections, in parties and also within governments – the so-called presidentialization of politics (Poguntke and Webb 2005). Given the rise of politicians who seem to transcend party structures, as well as the debate on the centrality of leaders, the fundamental challenge is for scholarship to fully incorporate leaders in political analysis. More recently, the field has been engaging in the measurement of ‘leaders’ and ‘leadership’ in a comparative and contextual way, especially in the electoral field (Aarts et al. 2013; Bittner 2011; Lobo and Curtice 2015). I have contributed to the measurement of the importance of leaders in a comparative context by seeking to integrate leaders – as party leaders – in the study of elections. I have tested the hypothesis that leaders of catch-all parties should be more important than leaders of mass-based parties for their voters (Lobo 2008). In addition, I have also researched the link between dealigned voters and leader effects, showing that indeed leaders matter more for those without a party identification (Lobo 2015). This latter research informs us about the phenomenon of personalization as studies show that there is a rising trend of voter dealignment (van Biezen et al. 2012).

The volumes reviewed in this article constitute an important step forward in the study of the role of leaders and personalities in democracies. This is due to the fact that they are all longitudinal. The initial studies on personalization were rarely able to draw on the systematic data contained in several of the books presented here. What really singles out the books reviewed in this article is that they are properly testing the idea of the growing importance of leaders.

With the exception of the book edited by Gianluca Passarelli (2015), all of the volumes reviewed here deal with the concept of personalization rather than the presidentialization of politics. ‘Personalization’, on the one hand, seeks to identify leaders not only as important, but increasingly important for political parties, both from the perspective of party organization as well as their electoral effects (Karvonen 2010). ‘Presidentialization’, on the other hand, is the process whereby parliamentary regimes are becoming more presidential in their actual practice – in government, in party and in elections – without, in most cases, changing their formal structure, that is, their regime-type (Mughan 2000; Poguntke and Webb 2005). Presidentialization therefore, encompasses personalization.

Given the fundamental challenge of understanding leaders’ place and growing importance in democratic polities, we will group the analysis in terms of parties and leaders (Cross and Pilet 2015 and Passarelli 2015), elections and leaders (Garzia 2014), and media and leaders (Langer 2011 and D’Arma 2015). After considering each volume’s merits, we will highlight what are the perceived consequences of a growing role for leaders in democracies. In the conclusion, we consider future directions of research that could be undertaken to advance the field.

Leaders and Parties

The volume by William Cross and Jean-Benoit Pilet, The Politics of Party Leadership (2015), makes use of a wealth of data on political leaders in 14 countries between 1965 and 2012, collected through the Comparative Study of Party Leaders (COSPAL) project. ¹ To my knowledge, this is the first time such a large data set on party leaders has been constituted, which allows the authors to ask very important questions on the extent to which parties are becoming personalized (leader-centric), and with what consequences for democracy. It should be welcomed because it signals for the first time a move towards a systematic and longitudinal comparative study of party leaders, which hitherto had been mostly analysed on a case-by-case format, even when the overall framework was comparative.

According to Cross and Pilet, analysing the politics of party leadership implies looking at leader selection, the dynamics of leader races, and leader termination. Most chapters use multivariate regression analyses, which allow for the systematic testing of hypotheses regarding one of these aspects.

How much change in the party leader selectorate has occurred between 1965 and 2012? This indicator is crucial for an understanding of the degree of personalization of politics because directly electing the leader by members or electors (primaries) is an indicator of a more leader-centric party. Yet, as the volume shows, parties are conservative institutions. Party leader selectorate changes have been relatively scarce. Indeed, most parties continue to select their leaders through locally chosen delegates (54 per cent) although 25 per cent do have selection via all party members. Therefore, importantly, despite worries about the personalization of parties, it is not the case that the direct election of the party leader has become widespread.

The first chapter, by Marco Lisi, André Freire and Oscar Barberá, tests the convergence hypothesis – that is, whether, regardless of the genetic model of party organization, political parties are adopting the direct election of the leader (Cross and Pilet 2015: 14), thus following the trend towards a personalization of politics (van Biezen et al. 2012; Katz and Mair 1995, 2009; Scarrow 2000). The authors find that there is a convergence to more directly elected leaders in all party families – with the exception of the Communists, who are least likely to have a direct election of party leader – when the data are considered on a decade-by-decade basis. However, in the last decade, there is no real difference between left- and right-wing political parties in this respect. The main difference is that the left started earlier, in the 1970s and 1980s, and the right followed in the 1990s and 2000s. These findings point to the fact that although party families seem to matter, change is affecting more or less all families, rather than pertaining to any one particular family.

This book’s use of an extensive data set and sophisticated techniques as well as its longitudinal scope make it a watershed in the study of the personalization of leaders for parties. It shows that, perhaps contrary to what some expect, the trend towards the direct election of the party leader is not overwhelming. In addition, it is not typical of either side of the ideological spectrum, or of a given party family. So, it is an overall modest party trend. This is confirmed in the chapter by Mihail Chiru, Sergiu Gherghina and Juan Rodriguez-Teruel which seeks to explain changes to leadership selection rules. Taken together, these two chapters point towards a convergence pattern which crosses ideological frontiers in terms of the direct election of the leader.

Overall, this book also tends to show that the impact which direct election of the leader has in terms of the internal dynamics is small. The competitiveness of leadership races increases (chapter by Ofer Kenig, Gideon Rahat and Or Tuttnauer); the direct election of leader does not seem to bring in more women leaders (chapter by Bram Wauters and Jean-Benoit Pilet), nor does it make a large difference for either renewal or termination.

What is perhaps lacking in Cross and Pilet’s volume is an overt strategy towards understanding what drives the personalization of parties – or rather the trend towards the direct election of the leader. Rather, Cross and Pilet’s approach aims to be more of a comprehensive study of the politics of party leadership. In each chapter, authors choose the most relevant dependent variable and independent variables and formulate hypotheses, which allows for the testing of multiple hypotheses in a thorough way. However, it also leads to some fragmentation and lack of an overall narrative on the (modest) trend towards the direct election of party leaders.

In his edited book, The Presidentialization of Political Parties, Gianluca Passarelli (2015) concurs with Cross and Pilet in affirming the conservative nature of party organizations, by placing party genetics centre-stage in explaining the degree of party presidentialization. The book’s aim is to understand why party presidentialization varies between countries. The presidentialization of parties in this context refers to the ‘greater autonomy [that has been given] to the leader, with great independence on crucial political topics, namely the electoral campaign, ministerial appointments and public policies’. The book’s main hypothesis builds on Samuels and Shugart’s (2010) work. They argue that party organization tends to mimic constitutional structure. Thus, to the extent that the constitutional structure separates executive and legislative origins and survival, parties will tend to be presidentialized. Passarelli (2015: 11) argues that the degree of party presidentialization, which depends on constitutional structure, will also depend on the party’s genetics, namely the original organizational characteristics of a party’s construction and development; the presence or absence at the party’s origin of an external sponsor; and the role of charisma in the party’s formation.

The book is structured in the following way: each chapter starts by illustrating the constitutional design and form of government of the selected case, thus identifying the degree of presidentialization which exists in each case. Then the authors identify party features that may inhibit or emphasize the effects of constitutional structure on parties’ presidentialization. The cases include Australia, Brazil, Chile, France, Germany, Italy, Japan, Poland, the US, Ukraine and the UK. They have been chosen due to their variation in constitutional structure. Each case is categorized, from the most presidentialized to the least presidentialized, namely presidential; president-parliamentary; premier-presidential; parliamentary. The genetic party categories vary from presidentialized/national to factional/local. The focus is on parties which have been in power between the 1990s and 2012.

The results tend to confirm the initial hypotheses: namely, the presidentialization of political parties is a phenomenon that must inevitably arise in presidential regimes, but that has also occurred in semi-presidential ones; also, the genetic characteristics of political parties function as an intervening variable capable of accentuating the opportunities offered by the institutions from a presidentialization perspective (Passarelli 2015: 236). The author further finds that in parliamentary systems, strict presidentialization does not occur. Nonetheless, there are signs of political personalization in all parliamentary cases considered. In Germany and Japan there is limited autonomy of the prime minister. Oreste Massari is sceptical about the degree of presidentialization possible in Italy because Berlusconi’s party, Forza Italia, despite being highly presidentialized, was ephemeral due to the lack of institutional and constitutional opportunities which could have offered lasting support to the party’s presidentialization generated by its genetic characteristics. Richard Hayton and Timothy Heppell argue that the UK is a case where personalization has occurred in a context of enduring parliamentarism.

This book is therefore important in that it seeks to develop not only David Samuels and Matthew Shugart’s (2010) work, but also the presidentialization thesis originally developed in 2005 by Thomas Poguntke and Paul Webb. It does so in an elegant form, where the genetic characteristics of parties are brought in as explanatory factors to distinguish between differences in presidential regimes, such as the US vs Chile, and semi-presidential regimes such as France and Poland.

However, it would have been useful if Passarelli had made an effort to quantify the organizational characteristics towards personalization to make the trends clearer. In addition, the book does not really account for changes in the speed towards personalization in each country, given the attention paid to the continuity of parties since their birth, rather than change. Passarelli (2015: 246) does highlight the direct election of party leaders as the most important party trait determining the level of personalization of parties, a sign not only ‘of an attempt to partially overcome the growing lack of democratic legitimacy, but also a deliberate initiative for a greater centralisation of power in the hands of central office and the party leader’. To single out this trait is particularly important, because it signals how quantitative research on the personalization of parties can progress, namely by focusing on the drivers of this trend, which is the focus of Cross and Pilet’s volume.

Leaders and Elections

The third book presented in this review, Diego Garzia’s Personalization of Politics and Electoral Change (2014), is ambitious not only theoretically, but also empirically. According to Garzia (2014: 2), ‘if the growing role of party leaders as drivers of partisanship is taken into account, and properly modelled in the voting equation, then their electoral effect emerges as much stronger than it has usually appeared’. The book compares three countries, Britain, the Netherlands and Germany, which were chosen because of their institutional differences as well as the availability of longitudinal data. It focuses on individual data from 1961 onwards.

The book’s premise goes beyond Passarelli’s. Garzia (2014: 19) argues that, ‘as a result of the process of party transformation, partisan loyalties have shifted accordingly from a mere reflection of previous socio-ideological identities to the result of individual attitudes towards more visible partisan objects – and in particular their leaders’. In such a context, it may well be that a party’s appeal has become increasingly shaped by its own leader’s image.

As a result, the author proposes an alternative scenario for the analysis of voting behaviour. Party leaders, instead of being seen as residuals in a socio-psychological model of voting, become drivers of partisan attachment at the individual level. In order to test that proposition, several things are needed. Namely, changing the conception of party identification to partisanship, drawing on literature which acknowledges that party identification is not static, but has several influences. If that is the case, and there is indeed endogeneity between a party and its leader, we can no longer take party identification as immune from leader effects.

Proper modelling – namely two-stage estimation and instrumental variables – allows the author to show the behavioural consequences of personalization. Namely, that leader effects are stronger than normally assumed, and that they have been increasing over time. This is the strongest critique to the consensus that leader effects have minimal impact on voting behaviour.

Garzia then makes party identification the dependent variable and compares a sociological model with a valence model to predict party identification. He shows that not only is the latter model more explicative of party identification than the sociological model, but that in Britain, leader effects have grown in their ability to explain partisanship.

The effect of leader voting in the three countries is measured using exogenized party identification measures. To be exogenous, variables must be caused by forces outside the system of equations and must not be correlated with the error terms. For partisanship, these standards are readily obtainable with socioeconomic variables. Then, partisanship (measured exogenously) and the leader barometer is added to the model, as well as standard variables which compose the socio-psychological model of voting behaviour. The result is that leaders become more important in the three countries concerned.

The author then employs a counterfactual strategy to evaluate the net effect of leaders on voting behaviour. He compares the actual electoral outcome with the (simulated) outcome of an election in which the main party leaders are seen equally favourably by voters. The findings from this analysis show that of the 20 elections under consideration, 10 do indeed present a potentially decisive effect for political leaders (as perceived and evaluated by voters) (Garzia 2014: 78).

This is, therefore, a book which takes seriously leader effects. The use of instrumental variables to limit endogeneity is a positive innovation that lends credence to the findings. Had panel data been available, they would improve the reliability of the findings, both for the construction of the instrumental variables, as well as making it possible to distinguish between attitudes (party identification, leader barometer) in t-1 and behaviour (voting) in t. Had leader traits been available it would be better to use them than leader barometers, which tend to be highly correlated with attitudes towards party and policies. Employing both panel data and leader traits would ensure greater exogeneity among the relevant variables. In the future, further refining of the methodology as well as the quality of the data available will be important to confirm this research.

Leaders and the Media

The previous three books pose something of a puzzle. The books by Cross and Pilet as well as by Passarelli provide a conservative view of the degree to which parties change. Whether this is because they are averse to change, or whether it is due to the fact that they are constrained by constitutional formats as well as their own origins, parties have not changed that much in the last decades. This does not mean that parties are not leader-centric. They may be so from their creation and also because they adapt to their constitutional framework (either presidential or semi-presidential). However, there does not seem to be a strong trend towards a personalization of parties in organizational terms. How, then, can we square this work with Garzia’s hypothesis that leaders are determining for the partisanship that electors develop? The answer may lie in the way in which leaders use the media to reach out to electors. Given the decline in party membership (Mair and van Biezen 2001), as well as the trust in political parties generally, parties and their organizations are perhaps playing a smaller role as intermediaries between citizens and politics. This intermediation is increasingly taken on by the media, which seem to play a very important role in the formation of political attitudes. It may be that the psychological turn towards leaders among electors derives from an increasing personalization of leaders in the media. The examples which we have given concerning the appearance of outsider politicians such as Trump, Berlusconi or Marcelo Rebelo de Sousa certainly seem to point to the heightened role the media can play in the personalization of politics.

Indeed, the books by Ana Inés Langer, The Personalisation of Politics in the UK (2011), and Alessandro D’Arma, Media and Politics in Contemporary Italy (2015), are in-depth studies which focus on the way in which the media have contributed to the personalization of politics in Britain and Italy. Langer’s volume is important on several levels. First, it has a historical approach and analyses the phenomenon between 1945 and 2009, thus taking seriously the longitudinal dimension of the personalization phenomenon. Second, it focuses on what the author designates ‘the politicisation of the private persona’ – that is, references in the media to the personal qualities and private lives of politicians. Specifically, it tries to measure the degree to which the ‘sense of pervasiveness of the personal’ is confirmed quantitatively.

The author performs a quantitative content analysis of articles from two newspapers, the Guardian and The Times. The data sample of 5,139 articles includes every article across two weeks in the first three years in office of each prime minister, creating a sample that is both representative of the normal coverage of prime ministers, and as comparable as possible. Langer looks first at references to leadership and personal qualities, followed by mentions of leaders’ personal lives – thus grounding empirically whether there has indeed been a trend towards politicization of the private persona (Langer 2011: 71). How does the author classify the non-political dimensions of politics? While acknowledging the difficulty in distinguishing between adjectives that characterize leadership vs personal qualities, the author proposes the following division: adjectives such as ‘competent, intelligent, strong, reliable, honest and experienced’ were taken as leadership qualities, while ‘nice, interesting, cool, fun/good-humoured, loving, family oriented’ were associated with personal qualities (Langer 2011: 82, figure 3.6).

The results show that ‘the overall presence of prime ministers in the newspapers has grown over time. However, the increase has not been systematic or substantial. The largest increase occurred in the 1960s’ (Langer 2011: 76). It is Harold Wilson, rather than Margaret Thatcher or Tony Blair, who receives the highest figures for several of the indicators of presidentialization.

The author argues that the most remarkable shift in recent decades is, however, the politicization of the private persona. Indeed, ‘the references to personal qualities are practically absent for any Prime Minister before John Major’. This indicates important changes in the definition of relevant character traits: it has broadened from an almost exclusive focus on skills and qualities directly linked to fitness to govern to providing ‘a more rounded picture of the overall personalities of prime ministers, as leaders and human beings’ (Langer 2011: 88). Yet Langer (2011: 111) shows that despite having increased in the period after Major, the politicization of the private persona is still a very residual aspect of the leaders’ image: ‘[the study] does not claim that it has become the most important dimension of party communication, branding or media reporting’.

Even for Tony Blair, ‘who is regarded as the paradigmatic case of the politicisation of the private persona only nine percent of the articles examined made reference to these issues’ (Langer 2011: 165). Yet we should not conclude that the change is insignificant. It is indicative of recent politics that assessments of the political leader qua human being now often play an important role in the interpretive narratives of leaders’ rise and fall in popularity (Langer 2011: 166). Not only that, but the leaders’ personal side has become an integral part of the leader’s political identity, and also of the ‘new’ party, in the case of Blair, as ‘his persona was crucial to changing what it meant to be Labour’ (Langer 2011: 167).

For Italy, Alessandro D’Arma (2015) analysed how media and politics have related, interacted and influenced one another in the last 20 years of Italian history, producing phenomena such as Silvio Berlusconi and Beppe Grillo This book is important, first, because it focuses on the causes of personalization, and in particular the way in which control of commercial television paved the way for a personalization of politics, and, second, because it clearly depicts a very bleak picture of the consequences of personalization.

D’Arma’s Media and Politics in Contemporary Italy is structured in the following way: Chapter 1 appraises three causes which have been advanced for Berlusconi’s personalization of Italian politics: first, his influential media ownership coupled with his media presence; second, Berlusconi’s dominance of Italian politics for 20 years stemmed from his ability to embody cultural values that he himself promoted earlier via his television channels; third, his success was due to his ability to master the rules of political communication in the era of televised politics. Chapter 2 focuses on the public television network, RAI, and its politicization before and after Berlusconi. Chapter 3 examines media policy in Italy. It recounts the circumstances that allowed Berlusconi to have a dominant position in commercial television in Italy, as well as the left-wing government’s failure in the 1990s to curb Berlusconi’s control of commercial television. Chapter 5 recounts the decline of political journalism since Berlusconi. Chapters 4 to 6 consider the changes to political journalism and the impact of the digital revolution for the personalization of politics. The prevalence of internet use in Italy has given rise to a new type of political party which again is highly personalized – the 5 Star Movement, led by comedian Beppe Grillo.

The book is exemplary for its singling out of the evolution of the media landscape as the main cause of the rise of Berlusconi. Complete deregulation in the 1980s allowed Berlusconi to dominate commercial television by the 1990s. Indeed, ‘nowhere else in Europe was a single private actor allowed to run three national terrestrial channels’ (D’Arma 2015, 45). Once Berlusconi decided to enter politics, creating his own party and using his media group for organizational purposes, research has shown that news coverage by his three television networks systematically favoured Forza Italia candidates (Marletti and Roncarolo 2000, in D’Arma 2015: 5). A wealth of research has underlined the bias that Mediaset, Berlusconi’s media group, gave to his party over the years (D’Arma, 2015, 6). In addition, once Berlusconi became prime minister, he also had substantial leverage over RAI, which had traditionally been controlled by the political parties (Chapter 2).

The form of personalization assumed by Berlusconi involved the creation of a personal party, which was completely identified with its leader and organizationally weak by design, as well as a very strong ‘politicisation of the private persona’, where Berlusconi sought to embody and represent the values disseminated through his commercial channels in the previous decades. D’Arma cites Ginsbourg, who characterizes Berlusconi’s television as deeply conformist, repetitive, uncritically consumer-oriented, bombarded by advertising, celebrating opulence, and promoting escapist dreams (D’Arma 2015: 8). According to Sergio Fabbrini, ‘Berlusconi’s persona was the message of Berlusconi’s personal party. He simplified the language of politics, and staged his body like a celebrity, thus becoming the champion of “pop politics” and blurring the traditional boundaries between politics and entertainment’ (D’Arma 2015: 15). Despite the resonance of these arguments, and the number of authors cited, it seems that there is no Italian study equivalent to Langer’s UK study, mapping the rise of the politicization of the private persona in Italy.

These two studies, both of which focus on the media, are quite different in their purpose. Langer sets out to measure personalization in the post-war period in Britain, whereas D’Arma is seeking to explore the causes and consequences of the personalization of the Second Italian Republic by looking at the control of television exercised by Berlusconi. They seem to diverge in the importance they attribute to the personalization of politics. Langer’s careful study points to the conclusion that the personalization of news was higher in Wilson’s premiership than at any time since, whereas D’Arma assumes from the beginning that Berlusconi marks a turning pointq in the personalization of politics in Italy. Thus, more comparative studies of other media are needed to reach more systematic agreement on this issue. Even Langer suggests that looking only at quality newspapers may have raised the bar too high for measuring the rise of personalization, and ignoring television is particularly problematic when studying personalization. These authors’ conclusions converge, however, in how they identify the novelty of the ‘politicization of the private persona’ which occurred both in Britain and in Italy, with Blair and Berlusconi respectively. Even if Langer concludes that the trend has not become dominant, with political dimensions being more important, it is still a new factor which has changed the overall leaders’ profile.

What are the consequences of the personalization of politics?

This review article has attempted to present the way in which studies of political leadership have contributed to our knowledge of personality politics. We turn now to the way in which these authors see the consequences and implications of their findings for the functioning of democracy.

Cross and Pilet conclude that its impact is relatively modest, namely it does not diversify the leader’s profile by making it more gender balanced, nor does it make party leader contests more competitive. In Garzia’s volume, a whole chapter is dedicated to the normative and empirical implications of putting leaders at the centre of the socio-psychological model of voting. After weighing the extant literature with a more or less positive view of the consequences of personalization, Garzia reflects on the way in which the mediatization of leadership, especially on television, has moved the public’s attention away from the leaders’ role as politicians to those of persons. This may have the ‘non-trivial effect of rendering political choices indistinguishable from consumer choices, but in a context in which choices are heavily influenced by emotions and tastes, rather than reflective judgement’ (Simon, cited in Garzia 2014: 86).

Langer admits that she focused on the ‘politicisation of the private persona’ because she was dissatisfied with the view that ‘the inclusion of the personal must be damaging to democracy’. However, her research led her to question the equally simplistic opposite view – that the humanizing of politicians may have a positive effect by bringing people closer to politics, ‘precisely because of the very political implications of the [politicization of the private persona]’. First, this is because the constant mediatization of the public persona makes it more likely for ‘the greater the chances for citizens to feel the leaders are falling short, […] because the combination of unremitting mediated visibility and press negativity is not exactly promising’ (Langer 2011: 172–3). Second, it is because the accountability of leaders may be compromised. ‘[I]f the personal became a predominant subject in the discussion of politics […] it could make it easier to bypass the examination of hard facts’ (Langer 2011: 173). So, for instance, Prime Minister David Cameron’s emotions about the death of his young son and his publicizing of the way the NHS treated the child well are not substitutes for an analysis of Conservative policy towards the health sector during his premiership. When such personalization of policy issues occurs, it serves to depoliticize issues, as if they depended only on personal choice, masking and declassifying ideological options.

Third, according to Langer, the politicization of the private persona also raises the issue of leader selection. Not only will people with less linear pasts be more reluctant to run for office, there is a gender bias in this trend. ‘Family life, sports and personal appearance are all minefields for women, and so is paradoxically, the expression of emotions, which are valued in men as a revelation of their softer side, but are often interpreted as weakness or lack of control in women’ (Langer 2011: 175).

The book by D’Arma is also overwhelmingly negative in its assessment of the consequences of the personalization of politics in Italy. First, the way in which politics and media fused during the Berlusconi era transformed political journalism, which has basically abandoned all pretence of impartiality (D’Arma 2015: ch. 5). Second, Berlusconi’s mark on Italian politics was so strong that new leaders who have since emerged have done so within the context of a personalization of politics. D’Arma cites Beppe Grillo as well as Matteo Renzi, the new left leader and former prime minister, as examples of this trend. Indeed, Renzi is often called the ‘Berlusconi of the Left’ (D’Arma 2015: 122). The drive towards celebrity politics, and the continued reinforcement of the link between politics and entertainment, may also have facilitated the emergence of a comedian as political leader such as Beppe Grillo.

Conclusion

This review started by acknowledging the emergence of the outsider politician, both in the US and Europe, stating that these developments pose challenges to the understanding of the role of leadership in democracies. The books analysed here try to cover different dimensions of the growing importance of leaders, namely for political parties, in electoral behaviour and in the media, in longitudinal terms.

Concerning parties, the research presents a case for moderate personalization of parties. According to Passarelli, parties may be personalized from birth and also as a result of adapting to their constitutional framework (either presidential or semi-presidential). However, there does not seem to be a strong trend towards a personalization of parties in organizational terms (Cross and Pilet 2015). These conservative conclusions run somewhat opposite to the volume by Garzia (2014), which argues forcefully for the growing importance of leaders in electoral behaviour. His research concurs with others which have also found for the importance of leaders for voting (Bittner 2011; Lobo and Curtice 2015).

Given the disjunction between the two dimensions of leader importance in parties and elections, it is important to search for other mechanisms which can become more important in personalizing electoral behaviour. The obvious candidate is the media, and we selected two books which deal with exactly the question of the importance of the media for the personalization of politics in the UK and Italy. Yet, Langer finds that personalization has not increased over time in British media. What is new is the inclusion of personal elements within the politician’s image. Still, these conclusions were solely based on high-quality newspapers and may have missed the wider picture in terms of television coverage as well as tabloids and the rise in personalization. D’Arma, in covering the Berlusconi years, is much more forceful in highlighting the media’s crucial importance for the personalization of Italian politics. As is often the case, the research does not all point in the same direction. However, it seems to me that this group of books suggest that the lack of substantial change at the party level does not necessarily mean that the personalization of politics is not occurring through other intermediaries, namely the media taken as a whole, and not only looking at quality newspapers.

The high quality of the studies surveyed raises a number of important questions, going forward. The first is benchmarking. The maturity of the field in measuring the importance of leaders begs the question of how we measure the importance of leaders relative to other indicators. Second, going beyond institutional contexts where leaders may become more important, it may be important to discuss which policies and policy outputs can explain personalization. D’Arma’s book is the only one in this collection that tries to make a connection between Berlusconi’s policies and the way in which they contributed to increasing his power. Third, the rapid change in the media landscape poses huge challenges in the measurement of the personalization of the media: how to incorporate and measure personalization at the level of the internet. Indeed, Langer (2011: 177) argues that ‘it would be important to combine soft and public affairs genres to enable us to explore the cross-fertilization across outlets, which is a key reason why there is such a sense of pervasiveness of the personal, which is not confirmed quantitatively’. Last, but certainly not least, the impact of personalization on the range and quality of representation should also be taken seriously.