SUCCESSIONS AND TRANSITIONS IN PARTIES AND GOVERNMENTS

The major study by William P. Cross and André Blais (2012) is devoted to the study of party leadership selection and removal in the five principal Anglophone parliamentary democracies (the UK, Ireland, Canada, Australia and New Zealand) between 1965 and 2008. Given the dearth of comparative studies in this field, this study is more than welcome.

The data for this ambitious undertaking come from an extensive review of contemporary and historical party statutes, of newspaper accounts of leadership contests and party rule changes, of leaders’ biographies as well as academic analyses. These primary and secondary materials are complemented by semi-structured interviews with approximately 50 party officials, senior party staff and principal political journalists.

One of the key findings of the related chapters on aspects of leadership selection and removal is that in all of the five Westminster democracies the parliamentary party does matter. Even though there has been a movement towards an expansion of the leadership selectorate (very often carried out after an electoral setback and while the party is in opposition), the parliamentary party tends to play an important role. Its power is felt even more strongly when it comes to removing a leader: in a majority of the parties covered in this study, the parliamentary party can even remove the leader completely if it wishes, and as the authors conclude, ‘leaders cannot survive indefinitely without significant support from the parliamentary colleagues regardless of whether they have the formal authority to remove him’ (Cross and Blais Reference Cross and Blais2012: 109).

As this comparative inquiry also reveals, leadership politics in parties has gathered pace: while the length of tenure varies across countries, it is declining everywhere, and very independent of the different rules governing leadership selection and removal (Cross and Blais Reference Cross and Blais2012: 93–4) – a development that most party officials interviewed by the authors attribute to the increased media and voter attention on the leader, and to the latter's increased vulnerability (Cross and Blais Reference Cross and Blais2012: 93–4). Also, irrespective of important differences between countries, most leaders do not leave voluntarily (Cross and Blais Reference Cross and Blais2012: 98).

Arguably the most fascinating findings are presented in Chapter 6 on running, getting elected and staying in office. The first thing we learn here is that ‘most leadership elections are not competitive. In the majority of the cases, there is no contest, as the sole candidate is crowned’ (Cross and Blais Reference Cross and Blais2012: 114), and while there have been slightly fewer coronations after 1985 than in the 20 years before, ‘there is no change in the number of candidates or in the closeness of the outcome’ (Cross and Blais Reference Cross and Blais2012: 115). Some important differences between countries and parties are worth noting: coronations are much more usual in Australia, New Zealand and the UK than in Canada and Ireland; everywhere, new parties are more likely to have coronations, while, ceteris paribus, opposition parties tend to have fewer coronations (Cross and Blais Reference Cross and Blais2012: 116).

Interestingly, there is no strong relationship between the size or degree of inclusiveness of the selectorate and the degree of competition in the leadership selection process (Cross and Blais Reference Cross and Blais2012: 118). Perhaps even more intriguing, leadership selection and removal rules do not appear to influence systematically the likelihood of choosing a younger person or a woman. Leaders elected by the full membership tend to be somewhat less experienced but, again, it is the parties that make more of a difference. Small opposition parties tend to have younger people at the top, which in the context of Westminster democracies can be explained by the fact that they are much less likely to be choosing a future prime minister (Cross and Blais Reference Cross and Blais2012: 119). Women candidates too seem to fare better in small parties. Challenging prominent assumptions about the positive correlation between a broad selectorate and the life expectancy of leaders, the authors find that leaders selected by a broad base do not have an easier or longer life but indeed appear to face a higher risk of having to depart early (Cross and Blais Reference Cross and Blais2012: 125).

Chapter 8, in which the authors look beyond their sample and review the features and developments in two non-Westminster democracies (Belgium and the US), mainly serves to demonstrate that the pressure to expand participation in selection is not confined to countries in the Westminster tradition. These cases also remind us that in terms of expanding the selectorate the Anglo parliamentary democracies have been followers rather than pioneers.

In their concluding chapter, Cross and Blais offer their recommendations for dealing with the key challenges of leadership politics in parties: For them, ‘the ideal leadership selectorate is one that includes both a party's parliamentary wing and its rank-and-file members’ (Cross and Blais Reference Cross and Blais2012: 174), which is mainly an acknowledgement of the special expertise of parliamentarians to assess the skills required to succeed at the job of party leader and possible prime minister. To Cross and Blais, the key reason why partisans from outside the parliamentary party should be given a formal voice in the party leadership selection is the institutional context of the Westminster parties (and other parties in parliamentary democracies), which is marked in particular by the indirect election of the prime minister (Cross and Blais Reference Cross and Blais2012: 181). This comes as a somewhat astonishing argument, as the authors previously explained the historical opening-up of the process for the selection of candidates in the US by the direct election of the American president. But, of course, empirical explanations and normative advice are two different things. While most recommendations by the authors come close to approving the status quo in the majority of the parties studied in this book, they take issue with the established rule that, with the exception of the Australian parties, there are no fixed terms for party leaders. This is – rightly – considered to be inconsistent with even the most basic democratic norms. Subjecting leaders to scheduled re-election votes could and would mark a distinctive feature of intraparty democracy. Needless to say, there would have been some room for other suggestions. In particular, some recommendations about how to increase the share of women leaders in political parties, or at least some acknowledgement of the importance of this issue, would have been expected and appreciated, especially as the authors highlight the low number of women party leaders as ‘perhaps the most stark finding with regards to gender’ (Cross and Blais Reference Cross and Blais2012: 121).

If there is one major item that many readers might have loved to learn a bit more about, it would appear to be the issue of ‘power versus authority’, whose special importance in understanding the politics of leadership selection and removal is recurrently highlighted by the authors (see, for example, Cross and Blais Reference Cross and Blais2012: 3, 8, 25, 109–11) but which remains somewhat underexplored even at the end. In Politics at the Centre this important topic is largely reduced to the blatant fact that those players within the party that do not have a formal say may still have major influence. Related questions worthy of some inquiry would be, for example, why some actors enjoying wide-ranging formal authority fail to transform this into power, and how exactly actors lacking any formal authority manage to win power and legitimize their notably successful attempts at shaping party fortunes.Footnote ²

The volume How Power Changes Hands, edited by Paul ‘t Hart and John Uhr (Reference ‘t Hart and Uhr2011a), shares the focus on the Anglo-Saxon democracies and elaborates on some of the issues dealt with by Cross and Blais. Starting from the study's subtitle, ‘Transitions and Succession in Government’, the editors set out with several useful definitions: while the term ‘transition’ is understood as ‘the arrival of an alternative set of hands ready to replace the existing team’, and more specifically ‘a change in the party or coalition that holds government’, ‘successions’ are referred to as ‘replacements of individual power-holders by others’, or ‘personnel changes in individual leadership positions within government (or political parties)’ (‘t Hart and Uhr Reference ‘t Hart and Uhr2011b: 2, 4).

In their overview, the editors rightly emphasize that their volume is ‘the first ever attempt to juxtapose both forms of ‘power changing hands’ and explore their antecedents and implications in one study’ (‘t Hart and Uhr Reference ‘t Hart and Uhr2011b: 6), but the result is only partly convincing, in particular in terms of coherence. The impressive line-up of contributors and their subjects of research, many of which are fascinating in themselves, do not really add up to a reasonably integrated volume. There is no framework that would serve to keep together the various chapters – which are devoted to such varied topics as conventions guiding caretaker governments in Australia and New Zealand (Simms Reference Simms2011), prime ministerial dismissals in the UK and New Zealand (Dowding and McLeay Reference Dowding and McLeay2011) and the afterlife of Australian prime ministers (Strangio Reference Strangio2011). Most chapters develop their own conceptual and theoretical agendas and therefore necessarily stand alone. Two genuinely theoretical contributions offering a specific conceptual framework – a chapter by Emina Subašić and Katherine J. Reynolds (Reference Subašić and Reynolds2011) on the social identity contexts and processes of transitions and successions and another one by Paul Atkins (Reference Atkins2011) on trust-building, learning and collaboration – are located at the back of the volume with no identifiable impact on any other contribution. In their conclusion the editors do their utmost to pull these distinct strands together, but to some extent this makes the volume's moderate degree of coherence only more strongly felt. As ‘t Hart and Uhr (Reference ‘t Hart and Uhr2011c: 239) correctly state, ‘the studies in this volume map out a broad terrain empirically as well as conceptually’ – neither more nor less.

That said, all the chapters gathered in this volume are solid enough, and some of them are more than that. The latter is particularly true for the chapter by Matthew Laing and Paul ‘t Hart (Reference Laing and ‘t Hart2011), which provides a comparative analysis of party leaders’ successions in 23 democracies. The authors' database comprises 515 cases of succession between 1945 and 2009, and their findings include quite a few counterintuitive results. To begin with, studied from a comparative perspective the Westminster systems are not marked by briefer life expectancies of party leaders. Neither are there significant differences in patterns of leadership longevity between two-party, multiparty or ‘two-and-a-half party’ systems (Laing and ‘t Hart Reference Laing and ‘t Hart2011: 114). The reported finding by Cross and Blais, that contemporary leaders serve for a far shorter period than their post-war counterparts, is confirmed for Laing and ‘t Hart's larger group of countries (Laing and ‘t Hart Reference Laing and ‘t Hart2011: 115). Further, the analysis suggests that the type of successor makes a difference in terms of success: while ‘loyalists’ are the most popular choice for a successor, ‘their success in getting chosen does not translate into success with the public. Instead, those with clear reform agendas – the “reformers” and the “revolutionaries” – seem to be the most successful types of leaders in terms of longevity and election results’ (Laing and ‘t Hart Reference Laing and ‘t Hart2011: 128). According to the findings presented here, there is also no strong correlation between the circumstances under which leaders come into office and their life in office: leaders gaining office through a high-conflict succession face only a moderately higher risk of losing office in the very short term than leaders emerging from more peaceful successions (Laing and ‘t Hart Reference Laing and ‘t Hart2011: 127).

As ‘t Hart and Uhr (Reference ‘t Hart and Uhr2011c: 244) highlight in their concluding chapter, a whole series of key questions for future research on their subject relate to the normative dimension of transitions and successions. It would indeed be useful to know more about what different modes of transition or succession are best – or considered best – for a democratic regime, and in what sense. As, in political reality, notions of legitimacy are effectively defined by the citizens in a given polity, no easy answers offering solutions to any type of regime should be expected. To date, there is not even a consensus about the relevant dimensions of a normative assessment of successions and transitions. The three aspects presented by Laing and ‘t Hart (Reference Laing and ‘t Hart2011: 129–30) as central normative dimensions of transitions and successions – their impact on policy, on the popularity of outgoing and incoming leaders, and on the legitimacy of the regime – would rather appear to represent three important areas for further empirical inquiry that could successfully link this emerging area of research to several of the more established subdisciplines of comparative politics. Another important aspect of transitions and successions, which is only cursorily touched upon in How Power Changes Hands, is the politics of communicating transitions and successions to the people. How leaders publicly explain or justify what actually happened would appear to have a major impact on both empirical and normative assessments, as well as on public perceptions, of political successions and transitions.

PUZZLES OF GOVERNMENT FORMATION

Most concepts of party government have little to say about government formation. The standard assumption is that the party winning the parliamentary election forms the next government and fills the key governing posts with its senior representatives. In this regard, with few exceptions – mostly from the family of Westminster democracies – party government has always been an illusion. In the large majority of established parliamentary democracies party government has effectively meant coalition government, with most new governments arising from protracted bargaining between parties (none of which is able to govern alone). As Hans Keman and Ferdinand Müller-Rommel (Reference Keman and Müller-Rommel2012: 3–4) observe in their introduction to a recent volume on party government in Europe that cannot be given more extensive coverage in this article, the rise of more complex party systems has not swept the major established parties from government but nevertheless increased the variety of coalition governments, many of which at least temporarily comprised new parties of the left and right.

Because elections in most parliamentary democracies do not tend to produce clear-cut winners that form single-party governments, understanding how election results eventually lead to a given government has rightly been acknowledged to be one of the truly major questions of comparative politics. This has been reflected in an impressive body of scholarly research on coalition building, with coalition theory representing one of the most advanced areas of theoretical research within the discipline. For all its theoretical sophistication, the explanatory power of established coalition theories, if judged in the light of real-world coalition politics, has remained limited. Indeed, fewer than half of all governing coalitions formed in Western Europe of the post-war period have been minimum-winning coalitions (coalitions enjoying the smallest possible majority support in parliament), traditionally being identified by different coalition theories as the most likely outcome of the competition between office-seeking parties. The office-seeking focus has long been complemented by a focus on policy preferences, but still many coalitions that are actually being formed defy coalition theory.

The important volume edited by Rudy B. Andeweg, Lieven De Winter and Patrick Dumont (Reference Andeweg, De Winter and Dumont2011) does not aspire to launch yet another general theory designed to explain the complex game of coalition building tout court. Rather, the editors and their contributors seek to advance our understanding of government formation in West European parliamentary democracies by revisiting specifically those cases that conventional theories of coalition building have been unable to predict or to explain. They do so by using a mix of methodological approaches. In a first step, those cases that deserve to be characterized as real puzzles of coalition formation in Western Europe between 1970 and 2005 are identified through a succinct statistical analysis. The cases eventually selected are all clearly deviant, in fact deviating by at least 65 per cent from the outcome predicted by established coalition theories. These cases – coalition formations from seven West European countries (Sweden, Norway, Austria, Spain, Finland, the Netherlands and BelgiumFootnote ³) – are then studied in great detail, with the goal of identifying those factors responsible for the deviant outcome. Methodologically, the authors proceed with ‘thick descriptions’ and the analytical tools of process tracing that, unlike traditional historical analysis, aim not only to provide the fullest possible understanding of a given case but also to arrive at generalizable explanations about the phenomena studied and to stimulate theoretical revisions and amendments.

Space does not permit a detailed reporting on the internal dynamics of these cases here, but even a more condensed overview of the key findings should give a fairly good impression of the major achievements of this joint venture. While the chosen methodological approach may somewhat bias the results towards overestimating the importance of agency, the case studies leave no doubt that the personal skills of the actors involved in the coalition-building game do indeed matter a great deal. The same is true for personal relations between key actors. As Rudy Andeweg (Reference Andeweg2011: 201) suggests in his insightful conclusion to the volume, the strong impact of the personal skills of the party leaders – or the lack of them – and the relations between actors on the government-building process appear to have been favoured by the exceptional complexity that marks all of the case studies gathered in the volume. Another lesson to be drawn from this intensive inquiry concerns the importance of the individual parties for the results of bargaining processes; it becomes particularly apparent in such multilevel contexts of coalition building as those observed in Belgium and Spain. As Andeweg (Reference Andeweg2011: 202) further notes, the case studies also suggest that the ‘electoral anxiety’ of the parties thinking about joining the government is no less important a factor in shaping the process of government formation than the expected benefits of governing (mainly office and/or policy) that continue to be at the centre of established coalition theories. Occasionally, this anxiety also manifests itself in parties urging their direct competitors to join as well, even if this will have to be paid for in terms of office and/or policy.

One other major factor brought to the fore by many of the case studies gathered here is the importance of informal norms that have been largely and systematically neglected by rational-choice based coalition theory. This is also one of the key themes of the chapter by Ilja van Beest (Reference van Beest2011) on psychological approaches to coalition formation, whose exact role for the field remains somewhat uncertain in this volume. In any case, the attempt to bring different disciplines and theories together and stimulate mutual inspiration and exchange is very welcome. Indeed, it seems that the importance of psychology is generally about to become more recognized in party scholarship, and not only in coalition research.Footnote ⁴ Both unwritten rules and psychology may lose their elusiveness if they are looked at through the lens of recent research on informal institutions which emphasizes the structural qualities of many soft variables and their impact on the behaviour and interaction of individual and collective actors (see Helmke and Levitsky Reference Helmke and Levitsky2004). That would, of course, require another paradigmatic transaction. While informal institutions, such as coalition committees, have long been at the centre of political research on coalition governance, there is no equivalent in research on the conditions and outcomes of coalition building.

COALITION AGREEMENTS, MINISTERS AND COALITION GOVERNANCE

The next book to be presented here focuses on precisely such informal structures that have come to shape coalition governance in many parliamentary democracies. In Coalition Government and Party Mandate, the Belgian political scientist Catherine Moury (Reference Moury2013) takes a careful look at what coalition agreements actually mean for decision making in West European parliamentary democracies, and the room for manoeuvre of cabinet ministers more specifically. While Jean Blondel's generous pre-publication endorsement – reproduced on the first page of the book – praises the author for having ‘provided us with the first true comparative empirical understanding of what goes on inside West European coalition governments’ is of course an exaggeration,Footnote ⁵ Moury's study marks nevertheless a remarkable achievement.

The key argument of this book, as well as its main result, is easily summarized in the author's own words: ‘coalition agreements are an efficient mechanism of coalition governance, and . . . cabinet members are effectively and consistently constrained by the document when making policy’ (Moury Reference Moury2013: 137). This main finding, plus a wealth of important details, is the result of an empirical analysis of recent cabinets in four West European countries (Germany, Belgium, the Netherlands and Italy). Methodologically, the study is based on a combination of approaches, including an empirical analysis of electoral manifestos, coalition treaties and cabinet decisions as well as interviews with about 50 ministers and 20 of their closest senior staff members. In a style that displays the typical strengths and weaknesses of a (carefully revised and amended) PhD dissertation – utmost terminological and analytical precision, and a slight propensity for redundancy – the author presents four case studiesFootnote ⁶ before offering detailed comparative assessments and conclusions.

Overall, the results provide ample empirical evidence that coalition agreements do matter in terms of policy, and that most ministers are indeed significantly constrained by them. In all four countries a large majority of the policy pledges from the coalition agreement are transferred into cabinet decisions and almost one-third of all cabinet decisions were precisely sketched out in the coalition agreement (Moury Reference Moury2013: 110–11). More than two-thirds of the respondents felt strongly compelled to implement the coalition agreement, and about one-third felt to some extent constrained to do so – with major differences between countries. Three-quarters of all Italian ministers feel only somewhat or not at all constrained by coalition agreements, with figures for the three other countries ranging between 21 per cent (Germany) and just 7 per cent (the Netherlands). Not surprisingly, Italian ministers also feel incomparably freer to propose to the cabinet new initiatives that are not in any way mentioned in the coalition agreement (Moury Reference Moury2013: 109). More than 50 per cent of all respondents considered limiting future conflict between the coalition parties to be the main purpose of the coalition agreement, with agenda setting, trust building among partners and communication with citizens considered important purposes by about one-third of respondents (Moury Reference Moury2013: 114). Particularly detailed and comprehensive coalition agreements were overwhelmingly explained by there being a lack of trust between the bargaining parties (Moury Reference Moury2013: 115).

As coalition agreements are not legally binding documents, their apparently strong impact on ministerial behaviour is in need of explanation. As Moury reports, there is a large group of mechanisms that effectively work to constrain ministers. In all four countries the cabinet is identified as a particularly powerful player that can stop a colleague's initiative. By contrast, when it comes to forcing a minister to act, it is the coalition committee, rather than the cabinet, that is identified by most respondents to be of primary importance.

Chapter 9 presents some ‘preliminary findings’ on the degree of collectivity and collegiality of the cabinet in the four countries. According to Moury, Belgian cabinets were the most collective ones (followed by Germany, the Netherlands and Italy, in that order), whereas Dutch cabinets were the most collegial (followed by the Italian centre-left cabinet chaired by Romani Prodi, Belgium and Germany). The fact that Italy under Berlusconi leaned more to the hierarchical end of the spectrum suggests that the prime minister is an exceptionally powerful factor in determining the degree of collegiality of a given cabinet. As to the degree of party control of the government,Footnote ⁷ which is also assessed in light of the previously presented case studies, Belgium and the Netherlands score considerably higher than Germany and Italy. These findings are interesting in themselves and their implications are not confined to the subject of coalition agreements and ministerial leeway. Within the context of this study the main conclusion drawn is that ‘cabinets that are collective, hierarchical and strongly controlled by parties (like Belgium) seem to be more likely to transfer the coalition agreement into cabinet actions than more segmented, collegial or loosely party-controlled cabinets (as in Italy)’ (Moury Reference Moury2013: 135).

In her brief conclusion, the author claims that many of her findings have ‘important implications for the quality of democracy’ (Moury Reference Moury2013: 139). She is certainly right, although this point is barely elaborated on, and at least some of the contentions made are somewhat questionable. For example, Moury (Reference Moury2013: 135) argues that ‘the fact that ministers are constrained by a document that is voted upon by parliament is certainly a good thing.’ This is not the case everywhere, however. In many countries, coalition agreements are approved by the extra-parliamentary party organizations of the coalition partners rather than by the parliamentary party groups or parliament as a whole. Also, Moury (Reference Moury2013: 139) states that ‘coalition agreements are virtuous because they allow package deals, and therefore significant change.’ The latter part of this contention may be right, but package deals tend to come at a high price in terms of transparency and accountability. A more extensive and balanced discussion of the normative dimensions of informal coalition governance is needed, but any future attempt is likely to draw valuable inspiration from Moury's work.

PARTY PATRONAGE AND THE CHANGING FEATURES OF PARTY GOVERNMENT

Realistic concepts of party government have never been confined to aspects of party competition, government formation and public policy making. The staffing of positions beyond the cabinet was early on identified as another crucial aspect of party government. Over the past years, the politics of party politicization of public institutions and society has developed into an independent area of research. In many studies, party patronage is understood as an electoral resource, as an aspect of clientelism (see, for example, Kitschelt and Wilkinson Reference Kitschelt and Wilkinson2006; Piattoni Reference Piattoni2001; Tomsa and Ufen Reference Tomsa and Ufen2013). The volume edited by Petr Kopecký, Peter Mair and Maria Spirova, Party Patronage and Party Government in European Democracies (2012), represents a different attempt at understanding party patronage. In contrast to most previous studies, the focus here is not on party patronage as a specific form of clientelism or brokerage but more on patronage as an organizational resource.

Party patronage as an organizational resource is essentially ‘a strategy to build parties’ organizational networks in the public and semi-public sphere’ (Kopecký and Mair Reference Kopecký and Mair2012a: 7). It serves to secure the parties a voice in the various policy-making forums that characterize modern multilevel governance systems and provides them with feedback from there. Many contemporary democratic regimes have experienced widespread processes of privatization and marketization, and a proliferation of agencies and non-majoritarian institutions that have been assigned responsibility for the formulation, implementation and regulation of public policy. ‘As governance becomes more dispersed, parties . . . have a greater need for an organizational network that can reach through to the different levels and arenas’ (Kopecký and Mair Reference Kopecký and Mair2012a: 13). And as ideological confrontations between parties as a previously dominant aspect of party competition have been largely set aside by a new focus on performance, the concrete strategies and agendas of patronage can be expected to reflect those altered features of party competition and governance.

Methodologically, this study draws on semi-structured, face-to-face interviews with nearly 650 experts (including academics, civil servants, journalists, non-governmental experts and party officials) conducted by country research teams. The experts chosen were not experts in researching patronage but rather had experience and expertise in at least one major policy area. The volume contains case studies of 15 European parliamentary democracies,Footnote ⁸ a selection that the editors wish to be seen as a reasonably diverse selection of countries (Northern and Southern, Western and Eastern European countries, large and small), rather than a more specifically composed sample (Kopecký and Spirova Reference Kopecký and Spirova2012: 23).

The subtitle of the first case study, by Oliver Treib (Reference Treib2012) writing on Austria, ‘From Reward to Control’, captures a leitmotiv of this extensive empirical inquiry into party patronage in old and new Europe. In many countries there has indeed been a substantial shift away from patronage as a reward towards using it as a means of control.Footnote ⁹ According to the aggregate data for all 15 countries provided by Kopecký and Mair (Reference Kopecký and Mair2012b: 362, Figure 18.1) in their conclusion, nearly 43 per cent of all respondents considered control the single most important motivation for parties to make patronage appointments, while only about 7 per cent mentioned reward. Needless to say, there are differences between countries, with regional strongholds of reward-oriented forms of patronage easily identified. Greece stands out as a showcase of heavily reward-oriented party patronage. The case studies on Spain, Bulgaria and Germany, too, point to the continuing importance of reward.

Given the strong emphasis that most parties now tend to put on control, it does not come as a surprise that the authors and their external experts identify professionalism as the single most important qualification of would-be appointees (88.1 per cent), followed by political allegiance (68.1 per cent) and personal allegiance (50.4 per cent) (see Kopecký and Mair Reference Kopecký and Mair2012b: 365, Figure 18.2). Alongside the increased professionalization of public administration (which creates a demand for professionally trained staff), the both more extensive and more intensive media scrutiny of public appointments seems to have shaped the observed strategies of party patronage. Whereas in the past longstanding party members were the natural beneficiaries of party patronage, more recent appointments tend to be made from a larger pool of candidates, many of whom have specialized professional qualifications but lack any institutional affiliation with the party. Further, there seems to have been a trend not only towards less obviously partisan patterns of party patronage. With few exceptions (such as Bulgaria and Spain especially), the very role of the parties in controlling patronage has been diminished in favour of individual leaders, in particular government ministers, taking more prominent roles (Kopecký and Mair Reference Kopecký and Mair2012b: 359).

The overall levels of party patronage in the 15 European parliamentary democracies covered show significant differences between countries. With an average score of 0.34 for all countries, Greece (0.62), Italy (0.49) and Austria (0.47) lead the field, whereas scores for countries such as the UK (0.09), the Netherlands (0.12) and Denmark (0.16) are conspicuously low. In many countries, the core civil service is the most patronage-ridden sector – with the UK, enjoying a score of 0.0, representing by far the most impressive exception to the rule. In most countries patronage is less pervasive in non-departmental agencies and commissions than in the executing institutions, while there are no similarly clear-cut patterns at the level of different policy areas (Kopecký and Mair Reference Kopecký and Mair2012b: 370–1, Tables 18.1 and 18.2).

The editors and their contributors provide several possible explanations for this empirical variation, such as the different types of governing coalitions, the minority/majority status of the government and, not least, the laws governing recruitment procedures for public offices. This does not really amount to a systematic explanation of the different manifestations of party patronage, though. As Kopecký and Mair (Reference Kopecký, Mair and Spirova2012: 372) acknowledge in their conclusion, many developments in the world of European party organizations remain immensely complex and inconclusive. The emerging patterns are, however, clear to see: As the parties themselves have gradually turned from firmly institutionalized organizations into organizational networks, their strategies of party patronage have come to focus on the creation and coordination of a network of policymakers who are expected to combine a respectable degree of loyalty with a maximum of professional expertise, with the overriding aim of improving the parties’ performance in government.