Non-partisan ministers in cabinets: who gets in and what do. they stand for? An approach and an application to Russia

Transcription

1 Non-partisan ministers in cabinets: who gets in and what do they stand for? An approach and an application to Russia Petra Schleiter Fellow and Tutor in Politics, St Hilda s College & Department of Politics and IR University of Oxford, Oxford OX4 1DY Tel: UK petra.schleiter@st-hildas.ox.ac.uk Abstract: A significant share of cabinet ministers worldwide do not act on behalf of parties, and to date, the tools available to political scientists to understand how non-partisan ministers are chosen and what governing strategies they represent remain poor. This paper synthesises the emerging literature on the causes of non-partisan ministerial appointments with elite studies work to develop expectations about the types of ministerial choices that politicians can be expected to make when they resort to nonpartisans. I test these expectations using an original dataset of 314 ministerial appointments in Russia. The results demonstrate that both contextual factors and the governing strategies of politicians have powerful effects on the types of ministers that are chosen. The findings contribute to a better understanding of cabinets that include significant shares of non-partisan ministers and have a series of implications for the literature on cabinet formation. Acknowledgements: I thank Paul Chaisty, Edward Morgan-Jones, and the participants of the Monday Seminar at St Antony s College (University of Oxford), for their suggestions and advice on earlier versions of the paper, and gratefully acknowledge the research assistance of Alisa M. Voznaya. 1

2 Non-partisan ministers in cabinets: who gets in and what do they stand for? An approach and an application to Russia Cabinet formation is the process by which politicians reduce the many potential governments that could result from voter choices to one particular governing team tasked with pursuing a specific set of policy directions. The attempt to understand this crucial choice has attracted major scholarly interest, and generated a large and sophisticated literature. This work has overwhelmingly been driven by an interest in cabinets that are based on party coalitions - their attributes, which parties are included, and how portfolios are divided among them - and has made great strides in answering these questions (for a good overview of the literature on government participation see Martin and Stevenson 2001; examples of work on portfolio allocation include Warwick and Druckman 2006; Carroll and Cox 2007; Verzichelli 2008). As a consequence, the study of cabinet formation, in particular in parliamentary democracies, but to a significant extent also in presidential democracies, has become virtually synonymous with coalition studies. Viewed through the lens of this literature Russian cabinets raise vexing questions. Between independence in 1991, and the end of Putin s tenure as president in 2008, not a single government minister was appointed on behalf of a governing party. How are ministers chosen when party affiliation and party attributes do not form the basis of ministerial choice? How can the governing strategies of politicians forming such cabinets be understood? Coalition studies approaches to cabinet formation are powerless to answer these questions, yet these are questions of real importance. The choices that politicians make about cabinet composition are no less consequential when they resort to non-partisans than when they form party based cabinets: When President Yeltsin appointed Yegor Gaidar s non-partisan government of shock reformers in 1992 he made a choice that would transform Russia economically and socially. This cabinet created tremendous opportunities for some, and imposed severe hardship on many. Similarly, when President Putin chose to appoint a fully nonpartisan cabinet under Prime Minister Fradkov in March 2004, despite the fact that the newly elected assembly was dominated by United Russia, a pro-presidential party with a near two-thirds majority he was making choices that fundamentally affected 2

3 the nature of political representation in Russia. To date, however, the tools available to political scientists to understand how non-partisan ministers are chosen, and what governing strategies they represent remain poor. If non-partisans in government accounted only for a trivial share of ministers, this gap in the literature would be of no great concern. However, the problem is not trivial and Russia, though extreme, is not sui generis. A third of a century after the start of the third wave of transitions that brought democracy to many countries in Latin America, and 20 years after the transitions in Eastern Europe, the universe of cabinets that require explanation has expanded well beyond Western Europe. As a range of studies show, party government is much less dominant outside Western Europe, and a significant share of the ministers around the world do not act on behalf of parties (Amorim Neto 2006; Amorim Neto and Strøm 2006; Schleiter and Morgan-Jones 2009; Samuels and Shugart forthcoming). Thus, while coalition studies have made great strides in accounting for cabinet formation in heavily party-dominated, parliamentary environments, and particularly in Western Europe, the understanding which political scientists have of cabinet formation beyond this narrow setting remains fragmentary. The challenge presented by the presence of significant shares of non-partisans in cabinets around the world is threefold. First, the critical predictors of government participation and portfolio allocation in standard coalition formation approaches - party attributes and institutions that structure bargaining between parties are on their own unable to account for the types of ministerial choices that politicians make when they resort to non-partisans. Accounting for such choices is likely to point up a range of influences on cabinet composition which coalition studies approaches have so far omitted. Second, the appointment of non-partisans cross-cuts traditional boundaries between the literature on cabinet formation in parliamentary and presidential regimes. Both types of democracies feature non-partisan ministers, albeit with different frequencies. The phenomenon of non-partisan ministers thus calls for an approach that can comprehend how politicians choose ministers in the different institutional settings that parliamentary, presidential (and indeed semi-presidential democracies) present. Third, governments that feature significant shares of non-partisans raise questions that cross the boundaries between work on democratic and non-democratic politics. Some 3

4 scholars argue that non-partisan ministerial appointments are a signal of nondemocratic government strategies, while others stress that the expertise which nonpartisans frequently bring to cabinets may help to legitimize democracy by enabling it to manage complex technical challenges. Neither possibility can be ruled out ex ante. Given that non-partisan ministers are most frequent outside Western Europe, where democracies are typically younger and often less consolidated, work in this area requires an approach that is not predicated on the existence of consolidated democracy but can distinguish between more or less representative and democratic cabinets. Indeed, these are some of the most important and interesting question to be asked of highly non-partisan cabinets. This paper examines how ministers are chosen when party affiliation and party attributes do not form the basis of ministerial choice and how the governing strategies of politicians forming such cabinets can be understood. In part one I begin by substantiating the claim that non-partisans account for a significant share of ministers cross-nationally by drawing together data on cabinets in Western Europe, Eastern Europe and Latin America. I then synthesise existing work on the appointment of non-partisan ministers, and argue that the types of non-partisans which politicians choose are partly shaped by the conditions that drive the initial resort to ministers who are unaffiliated with parties, and partly by the governing strategies of the politicians in question. In part two of the paper, I trade breadth for depth and apply the approach developed in part one to the analysis of the ministerial choices of Russia s first two presidents, Boris Yeltsin and Vladimir Putin, using an original dataset of 314 ministerial appointments from 1991 to The paper demonstrates that controlling for context it is possible to draw inferences about the governing strategies that politicians are pursuing from the types of nonpartisans they choose. It thereby lays the basis for a better and more nuanced understanding of cabinets that include significant shares of non-partisan ministers, which has implications for future work on cabinet formation. Amongst other things, the paper s results challenge the literature on cabinet formation to develop empirically and theoretically in the direction of assimilating the appointment of non-partisan ministers. The paper makes clear that a more contextualised understanding of government formation - which takes account of the varying constitutional frameworks 4

5 and policy challenges that confront politicians and of the governing strategies that they pursue - is likely to be critical to further progress in the field. Part 1: Literature, the salience of non-partisan ministers, approaches and expectations Work on cabinet formation in the parliamentary regimes of Western Europe, is largely based on the premise that in a parliamentary context, the most interesting and important questions about cabinet formation reduce to questions about the formation of coalition (rather than single party) cabinets. Cabinet posts held by non-partisans are typically not considered by this work and are excluded from empirical analysis. The literature on presidential cabinets, too, has given most attention to the party-cabinet relationship, focussing on the question how frequent coalition cabinets are in presidential regimes (Cheibub Figueiredo and Limongi 2000; Cheibub, Przeworski et al. 2004; Cheibub 2007), and how government participation is shaped by party attributes in those countries in which cabinets are based on coalitions (Altman 2000). Apart from a concern with the rise of tecnicos (technical experts, who rise to positions of political power) in Latin America little systematic attention has been given to nonpartisan ministerial appointments until very recently (Centeno and Silva 1998). The work of Amorim Neto altered the research agendas with respect to both Latin American and European cabinets. In a comparative study of presidential governments in 13 countries in the Americas, Amorim Neto documents the frequency of nonpartisan ministerial appointments in presidential cabinets and argues that the choice between non-partisan and partisan ministers is centrally shaped by a president s legislative strategy (Amorim Neto 2006). A second joint paper with Strøm, demonstrates the prevalence of non-partisans in European cabinets, too, and accounts for their appointment through the influence of presidential heads of state on cabinet formation (Amorim Neto and Strøm 2006). Schleiter and Morgan-Jones (2009) build on this work. They establish that non-partisans do not just control sizable numbers of government posts in East and Western Europe, but also critical portfolios including the premiership, interior affairs and foreign affairs, and show that the frequency of such appointments is shaped by the manner which Europe s semi-presidential and 5

6 parliamentary regimes structure the principal-agent relationships between voters, elected politicians and governments. Together these works make two contributions: First, by documenting the frequency and importance of non-partisan ministerial appointments, they demonstrate a major gap in the discipline s current understanding of cabinet composition. Second, they develop and test accounts which explain why politicians resort to non-partisan ministers. In doing so these papers give rise to a raft of new questions, and set new directions for research on government formation. Perhaps the most pressing questions raised by this research are how politicians choose what type of non-partisan ministers to appoint, and how the governing strategies of politicians who resort to non-partisan cabinet appointees can be understood. These are the questions addressed in this paper. I begin by drawing together comparative data on the frequency of non-partisan appointments which demonstrates the substantive importance of these questions and underscores the claim made in the introduction that Russia is extreme but not sui generis. I then synthesise existing approaches to generate a set of expectations about the types of non-partisans that politicians are likely to choose under a range of contextual circumstances and governing strategies. These expectations are applied in part two of the paper to the analysis of ministerial appointments in Russia between 1992 and Cross-national evidence: The salience of non-partisans ministers in Latin America, East and Western Europe Once the study of cabinet composition is extended beyond Western Europe, where the coalition studies approach to government composition has its roots, it becomes evident that the appointment of non-partisan ministers is a critical feature of a large number of cabinets. Figure 1 draws together the data collected by Schleiter and Morgan-Jones (2009), and Amorim Neto (2006), which detail the share of nonpartisan ministers in the cabinets of 41 parliamentary, semi-presidential and presidential democracies in East and Western Europe and the Americas. [Figure 1 about here] 6

7 As the figure makes clear, Russia s patterns of cabinet composition are extreme when compared to the group of semi-presidential regimes among which it counts, but not unique. Non-partisan ministers feature routinely in cabinets and on average, between 5 and 27 per cent of the cabinet appointments in parliamentary, semi-presidential and presidential regimes are left unexplained by approaches to cabinet formation which predominantly focus on party coalitions these numbers reach maxima of 65 and 100 per cent in some presidential and semi-presidential regimes. Without tools to understand these non-partisan appointments, the discipline s understanding of cabinet composition remains partial and inadequate. In sum, the patters of cabinet composition which Figure 1 summarizes constitute prima facie evidence of a need to engage with the question how politicians choose non-partisan ministers and what governing strategies such appointments represent. Generating expectations from existing approaches Non-partisan ministers are defined as ministers who do not act on behalf of a party in government. They may have known party preferences and even experience of serving a party, but they serve in government in a personal capacity, they are not representatives of a party, and their appointment does not constitute party participation in government. This section draws out expectations about who gets into government when politicians resort to non-partisans. My approach is predicated on the assumption that cabinet appointments are some of the most important appointments that politicians make because they offer a powerful tool to manage the policy process. The role of ministers in the policy process is to advise on policy alternatives and implementation strategies (Laver 2006: 125), liaise with constituencies and stakeholders and secure administrative and political support for policy implementation (Fenno 1958: ). From a politician s perspective the selection of ministers who will handle parts of the policy process on their behalf is a delegation problem. And I argue that the kinds of skills, experience, knowledge, and outlook that politicians value in the ministers they appoint depend in part on the institutional, political and economic environment that they are faced with, and in part on their own preferences and governing strategies. Two separate literatures help to understand how contextual factors and governing strategies shape the ministerial 7

8 choices of politicians: the work introduced above, which examines the causes of nonpartisan ministerial appointments, and work on political elites. I draw on both literatures in this section. a. Contextual factors and the literature on the causes of non-partisan appointments The work introduced above by Amorim Neto 2006, Amorim Neto and Strom 2006, and Schleiter and Morgan Jones 2009 makes clear that the institutional, political and economic context within which politicians must work affects the skills and qualities that politicians are likely to seek in their ministers. Institutional context: Separation of powers and purpose One of the most striking patterns to emerge from figure 1 and the work of Amorim Neto (2006), Schleiter and Morgan-Jones (2009), and Samuels and Shugart (forthcoming) is that cabinet partisanship correlates strongly with democratic regime type. My argument in this section is that institutions do not just shape the frequency with which politicians rely on non-partisans, but also the attributes that they are likely to seek in nonpartisans to ensure that the cabinet s incentives are compatible with their own. Specifically, constitutions affect what value politicians are likely to place on personal loyalty in constructing their cabinets. The reason for this is simple and relates to the degree to which different types of democracies fuse or separate the power and purpose of the executive and legislative branches of power. In parliamentary regimes, powers are fused and government is the agent of the legislature alone, which can depose its cabinet through a vote of no confidence (Müller, Bergman et al. 2003: 13). This gives assembly parties undiluted control over the allocation of government portfolios. The result is tight party control over government posts, which serves both the party in parliament and its agents in cabinet because their fates are mutually dependent given the fusion of executive and legislative power and their joint accountability in parliamentary elections (Schleiter and Morgan-Jones 2009: check page number). 8

9 In presidential democracies, by contrast, cabinets act on behalf of the president to whom they are accountable. Because legislators and presidents are separately elected and accountable in different constituencies, they must typically be responsive to different sets of voter concerns. As a result, the tensions between a president s mandate and that of his or her legislative party can be considerable (Samuels and Shugart 2006), and presidents cannot necessarily achieve high incentive compatibility by composing their cabinet of members of their party (Schleiter and Morgan-Jones 2009). Consequently, they are more likely to include loyal non-partisans in cabinet, making the party-cabinet relationship in presidential democracies significantly weaker than in parliamentary democracies (Samuels and Shugart forthcoming). In semi-presidential democracies, which are characterized by a popularly elected, fixed term president [who] exists alongside a prime minister and cabinet responsible to parliament (Elgie 1999: 13), the cabinet-party relationship lies between these two extremes: While the government s dependence for its survival on assembly confidence ensures that the party-government relationship under semipresidentialism is stronger than in presidential democracies, the influence of the popularly elected president on cabinet appointments ensures that the partygovernment link is weaker than under parliamentarism (Amorim Neto and Strøm 2006; Schleiter and Morgan-Jones 2009). In sum, the institutional context affects the means by which politicians can achieve cabinets which have incentives compatible with their own: While long and loyal party service is a powerful tool to ensure that ministers in parliamentary cabinets will act faithfully on behalf of the parties that appoint them, co-partisanship is a less powerful tool for presidents to ensure that the cabinet will share their aims, and they are therefore more likely to value personal loyalty as a critical attribute in their cabinet ministers. The first expectation that can be derived from work that comments on regime type effects on the type of minister that politicians are likely to appoint, then, is that: Hypothesis 1: The greater the separation of powers and purpose in a democracy, the greater the share of non-partisan ministers selected for personal loyalty to the president (all expectations are formulated subject to ceteris paribus conditions). 9

10 Political context: The assembly Whether their purposes coincide more or less with those of the assembly, presidents may have to work with legislators to achieve statutory legislation (and in semi-presidential regimes additionally to ensure the survival of governments). As existing studies point out, cabinet appointments are a critical tool for presidents to cement legislative coalitions (Amorim Neto 2006), and in semi-presidential regimes, to construct assembly coalitions that will support the survival of the cabinet in office (Schleiter and Morgan-Jones 2009). Yet, the opportunities to build such coalitions vary with assembly composition and typically rise with the size of the president s legislative support (Amorim Neto 2006), and the ability of assembly parties to act collectively (Schleiter and Morgan-Jones 2009). Clearly the most direct way to build such coalitions is to appoint partisan ministers who represent their parties in cabinet. But even when other factors lead politicians to prefer non-partisans, they may wish to draft legislative and party political expertise into their cabinets, and their incentives to do so should vary with assembly composition. The two expectations that can be derived from work that examines the impact of assembly composition on the type of non-partisans recruited into cabinet are that: Hypothesis 2: The greater the president s legislative contingent the greater the share of ministers with experience of legislative work and with links to propresidential parties. Hypothesis 3: The greater the assembly s ability to act collectively, the greater the share of ministers with legislative and party experience. Policy challenges Finally, policy challenges have been shown to affect cabinet partisanship, and again I argue that they also affect the qualities that politicians are likely to seek in their ministers. Technically complex policy challenges put a premium on technical expertise and enhance the value that politicians attach to technical training, such as PhDs and independent research careers, as compared to partisanship in forming the government (Bermeo 2003: 216; Amorim Neto and Strøm 2006). The Latin American debt crisis, for instance, was associated with the influx of tecnicos into governments. This demand for technical expertise was enhanced by the need to signal technical competence to outside actors in international financial institutions, 10

11 Washington and elsewhere (Schneider 1998: 85-92). The expectation that can be derived from this work is that Hypothesis 4: Technically complex policy challenges are likely to raise the share of non-partisan ministers in government who are technical experts. These three types of factors, constitutions, assembly composition and policy challenges shape the contextual constraints that politicians face as they choose their cabinet ministers. But within the range of options defined by the structural context, politicians clearly make deliberate choices to shape cabinet composition in a manner that reflects their specific approach, and it is the impact of governing strategies on ministerial choice that I turn to next. b. Governing strategies: legislative strategies and democratic preferences The governing strategies that politicians choose can aim for greater or lesser links with other political actors, and they can be more or less democratic. My argument in this section is that the type of skill and experience that politicians seek in non-partisan ministers is likely to be conditioned by their governing strategies which may vary along two dimensions: their choice of legislative strategies (in the case of presidents), and the degree of their commitment to democratic as opposed to non-democratic methods of exercising political power. Legislative strategies Amorim Neto (2006) first demonstrated that the choice of legislative strategy can be expected to have an impact on cabinet composition. For parliamentary politicians there is little choice: Parliamentary governments must typically work through statutes to achieve their aims, and must therefore construct legislative majorities, which often coincide with the parliamentary coalitions that keep them in office. However, to the extent that presidential and semi-presidential constitutions make cabinets less reliant on assembly support for their legislative success, presidents may have a choice between statutory and executive prerogative based legislative strategies. Seeking policy goals by means of statutes requires passage through the standard legislative process (Amorim Neto 2006: 416), and 11

12 presidents who adopt a statutory strategy are likely to use cabinet appointments to build the requisite legislative coalitions and to heed the views and interests of representatives in the assembly (Amorim Neto 2006: 417). Even when other factors push politicians to rely on non-partisan ministers, presidents who place a high value on enacting policy through statute may be expected to appoint ministers who have legislative expertise and experience of the parties that are represented in the assembly, so as to manage the legislative part of the policy process. Alternatively, presidents whose legislative powers are extensive may choose to govern by executive prerogative, that is decree and veto, and are less likely to use cabinet appointments to build legislative coalitions or links with the assembly to manage the legislative process (Amorim Neto 2006). The expectation that can be derived from Amorim Neto s work, then, is: Hypothesis 5: Strategies that favour legislation by statute (by contrast to executive prerogative based strategies) are likely to increase the share of ministers with parliamentary and party expertise. Non-democratic governing strategies At times, however, it is not contextual factors, or the reliance on executive prerogative, but authoritarian inclinations that lead to the appointment of cabinets with too much autonomy from other political actors and too much autonomy from voters as well, [so that] they harm the credibility of party systems and, indeed, the whole electoral process (Bermeo 2003: 222). Elite studies work that has compared cabinet composition under authoritarian and democratic rule shows that authoritarian politicians typically curtail the representative function of cabinets so that the cabinet becomes a predominantly administrative institution (Taveres De Almeida, Costa Pinto and Bermeo 2003). Cabinets formed by politicians who incline toward authoritarian rather than democratic means of governing are likely to substitute a heavy reliance on civil service and administrative experience for links with representative institutions, and to put more emphasis on the use of repression so that ministers with career backgrounds in law enforcement and the military tend to feature more heavily in the cabinet (Bermeo 2003: 208-9, ). This, then, is the last of the expectations that I derive from the literature. 12

13 Hypothesis 6: Authoritarian governing strategies are likely to raise the share of non-partisan ministers with administrative, military and law-enforcement backgrounds in cabinet. In sum, the expectations that can be derived from the existing literature indicate that two broad sets of factors are likely to shape cabinet composition when politicians reach for non-partisans, structural context and governing strategies. Part 2: Non-partisan ministerial appointments in Russia. Who gets in and what do they stand for? In this section I apply the expectations developed above to Russian cabinets formed between independence in December 1991 and the end of the Putin presidency in The rationale for examining Russia in detail is twofold. First, its cabinets so thoroughly escape comprehension through the coalition studies literature on government formation that it promises to give special leverage in pointing up factors which structure cabinet composition, but which that literature has to date overlooked. Second, building on the existing literature, my expectation is that contextual factors and the governing strategies of politicians are likely to have conspired to produce this outcome, and focussing on a single country allows me to trace the causal links between contextual factors, governing strategies, and government composition at a level of detail, and through a richer range of sources than would be feasible in a crossnational study. The Russia-specific version of the general expectations outlined above is that the nature of its constitution, which is semi-presidential, with an exceptionally powerful presidency (Schleiter and Morgan-Jones 2008: ), can be expected to give rise to a significant degree of reliance on non-partisan ministers, appointed for their loyalty to the president (H1). But by any measure, the fully non-partisan governments that have characterized the country are extreme. Why? Given the approach outlined above, the technically complex policy challenges involved in launching a transition to the market, and managing the prolonged transitional crisis which followed up to 1999 would be expected to lead to the recruitment of high levels of technical experts as 13

14 ministers, but once the economy began to grow in 1999, that pressure should have subsided (H4). Similarly, high initial party system fragmentation and Yeltsin s status as a president with minority legislative support made coalition building in the assembly, and a statute-based approach to policy making difficult which generated weak incentives for the recruitment of governments with links to the legislature and assembly parties. But after 1999 a pro-putin legislative coalition formed, and the 2003 Duma elections a pro-presidential majority party in an assembly with low levels of party system fragmentation. These contextual changes opened the door to strengthened party and parliamentary links to the government (H2 and H3). In responding to this institutional, political and economic context, politicians will have been guided by their governing strategies. A president with a preference for a statutory strategy of implementing his policies would be expected to build links between the assembly and the government through the recruitment of ministers who have standing with parliamentary parties, and parliamentary experience, even when confronted with a fragmented assembly (H5). By contrast, presidents relying on executive prerogative and authoritarian governing strategies would be expected to have formed cabinets that curtailed links with representative institutions, but only an authoritarian president should have based his cabinets predominantly on ministers with administrative expertise, and enhanced the repressive capabilities of the cabinet by recruiting individuals with military and law enforcement expertise (H6). In applying these expectations to Russia, I proceed in two steps. First, I probe the expectations regarding contextual factors - the constitution, economic crisis, changing levels of parliamentary fragmentation and parliamentary support for the president - which can be readily observed. Second, I examine the expectations regarding the impact of presidential governing strategies, on the cabinet. Because governing strategies are not readily observable ex ante, I proceed in reverse order and apply the expectations developed above to the cabinet appointments that the two presidents made given the contextual influences, to identify their governing strategies. To probe the reliability of my inferences about governing strategies, I examine to what extent they are consistent with independently observable implications of the two presidents approach to governing, including decree and statute use, and the quality of democracy. 14

15 Data and Method The dataset on which this section of the paper is based is original and covers the governments appointed by presidents Boris Yeltsin (from independence to 1999) and Vladimir Putin ( ). I record the professional backgrounds of all members of the government with full ministerial status. Government membership is registered for each new government, at the time of the government s appointment. A new government is counted with every constitutionally required government change (after presidential elections), change of prime minister, and turnover of at least fifty per cent of the cabinet members. In all, the data comprise information on the career histories of 314 ministers in 12 cabinets, 85 relate to the 4 governments formed by Putin during his presidency, while 229 relate to the 8 governments that Yeltsin formed. Table 2 summarizes the information on the governments, their appointment dates, and the number of ministers that they comprise. [Table 1 about here] Since I am interested in capturing that part of an individual s career history, which is most relevant to his or her promotion to ministerial office, I record careers in the five years that precede the ministerial appointment so that the skills and knowledge relating to those positions are up to date, and contacts active. The only exception to this rule are loyalty links with the president. Where the career path of a minister crossed that of a president in the past, such links were recorded without time limit. The career backgrounds of ministerial appointees typically comprise a range of skills, experience, and attributes, the combination of which makes them valuable as members of government. For example, the career of Aleksandr Shokhin, who served as Deputy Prime Minister and Minister for the Economy in Chernomyrdin s (January 1994) government, combined technical expertise (a PhD in Economics and experience of working in academia) with experience of running for and winning a parliamentary seat in the 1993 State Duma for the pro-presidential Party of Russian Unity and Accord (PRES), and with experience of having worked in the Soviet civil service as well as previous service as a Russian minister for Labour and Deputy Prime Minister. To capture the range of skills and experience that politicians are recruiting into government through their ministers adequately, the data records these multiple attributes. 15

16 As a consequence, the profile of ministers and cabinets can vary along a range of dimensions which may overlap. Cabinets in which technical expertise is more prevalent are therefore not necessarily cabinets in which parliamentary expertise or party experience is less prevalent, because ministers may combine technical expertise with parliamentary or party experience. Because ministers have multiple overlapping career backgrounds, regression frameworks do not lend themselves easily to the analysis of this type of data. For example, although the appointment of ministers with technical or parliamentary expertise can individually be characterized as dichotomous outcomes, and would in principle be amenable to a multinomial logit approach, these two categories may overlap ministers appointed may have technical and parliamentary expertise, or neither. Once the analysis is extended to the number of categories of interest here (e.g. loyalty to the president, parliamentary expertise, experience of a pro-presidential party or opposition party, administrative experience, military or KGB experience), and any of their combinations, the possible outcomes proliferate, and a multinomial logit analysis with just 314 observations is no longer feasible. Given these challenges I choose to keep the analysis as simple as possible, and rely on descriptive statistics to chart the nature of Russian ministerial appointments along the different dimensions of interest here. Accounts by specialists of Russian politics Before turning to the analysis of the data, a brief overview of the trends charted by experts of Russian politics is in order. The composition and change of political elites in Russia has attracted significant attention from experts of Russian politics. Studies tend to focus on the changing prevalence of a single attribute of political elites such as administrative expertise, military background, or loyalty to the president. This body of work has been characterized by debates about the meaning of various features of the Russian elite. Throughout the Yeltsin era, the debate was dominated by the questions of continuity and change. While some authors argued that continuity between Soviet and Russian elites was high, and boded ill for democracy (Hanley, Yershova et al. 1995; Kryshtanovskaya and White 1996), others pointed to significant turnover since the collapse of the USSR (Lane and Ross 1998; Rivera 2000), and described the influx of technical experts in particular into the elite. One set of authors describes the Yeltsin elite as follows: Under Yeltsin a rather different elite came to power, 16

17 recruited from a wide variety of social groups and for the most part without any serious management experience. Many were politically inexperienced academics. (Kryshtanovskaya and White 2003: 291). Since Putin s victory in the 2000 presidential elections, a range of changes among the incumbents of top political positions have attracted attention. Observers have noted a rise in loyalty based appointments, that is the influx of former colleagues of Putin in St Petersburg (the petertsy), and the KGB into positions of political power (Petrov and Slider 2003) and the very limited party-government links (Wilson 2006: 336). But particular interest has focussed on the influx of siloviki individuals with military and law enforcement backgrounds into the political elite. However, again, agreement as to the nature and meaning of these trends has proved elusive, as analysts have debated whether the influx of siloviki into positions of power marks the rise of a new authoritarian and militarized regime in Russia (Kryshtanovskaya and White 2003), or whether the diversity of outlook and background among those labelled as siloviki makes it impossible to draw such conclusions (Renz 2006; Rivera and Rivera 2006). In sum, the ongoing debates highlight that among experts of Russian politics, too, the understanding of why individuals with particular backgrounds are appointed to top political positions, and how to interpret the meaning of such appointments in terms of ruling strategies, remains limited. Who gets in? I now turn to the analysis of the appointments, which presidents Yeltsin and Putin made to their cabinets. The analysis here differs from the elite studies work on Russia reviewed above in two respects. First, it compares the relative prevalence not just of one type of background, but of multiple types of ministerial experience in government over time, and second, it is driven by a range of expectations which set out how context and governing strategies might shape the ministerial choices of Russia s presidents. I begin by examining how the two presidents responded to the contextual factors they were confronted with. Institutional framework The first of those is the constitutional framework, which, though semi-presidential, provided for a powerful popularly elected 17

18 presidency. While the focus on Russia alone does not allow me to test hypothesis 1, I can observe whether cabinet appointment patterns in Russia are congruent with it, and as expected, Yeltsin and Putin responded to the difficulties that popularly elected presidents face in constructing incentive compatible cabinets by appointing ministers as part of the cabinet mix to whom they had loyalty links. But as sub-table 2a makes clear, there were marked and statistically significant differences in the two presidents reliance on loyalty in order to build cabinets which would share their aims. While just 8 per cent of Yeltin s ministers had career connections with the president, over a third of Putin s ministers had loyalty links to the president. The substantive significance of this difference must be assessed in the context of the two presidents legislative support, and I return to this question below. But the constitution was not the only contextual factor that the two presidents were responding to, and other elements of the context, most notably the economic situation, parliamentary fragmentation and the size of the president s legislative support, changed over time, and constrained the two presidents ministerial choices to different degrees. Economic transition and crisis The task of launching Russia s transition from a planned economy to the market, and the severe recession that accompanied that transformation during the 1990s posed complex technical challenges and had, as expected, a sizable and statistically significant effect on the recruitment of technical expertise into the government (H4). Russia experienced a prolonged economic crisis in the 1990s, during the Yeltsin presidency, from which it only emerged at the very end of his period in office. As the upper part of sub-table 2b shows, while the economy was in crisis, the share of ministers with high levels of technical expertise, that is PhDs and/or academic and research careers averaged 52 per cent. Once the economy began to grow in1999, the demand for technical expertise became a much less powerful driver of cabinet composition, and the share of technical experts in government dropped to an average of 33 per cent under Yeltsin. Did the two presidents have differing preferences for the inclusion of technical experts in government? The answer, as the lower half of sub-table 2b suggests, is no. Since Putin never presided over a contracting economy, the two presidents appointment strategies can only be compared once the constraints of managing the economic crisis 18

19 had dropped away. Under conditions of growth, they appointed roughly equal shares of 33 per cent and 27 per cent technical experts respectively. As the t-statistic indicates, there is no significant difference between these two numbers. In sum, the economic transition and accompanying crisis proved to be a powerful constraint on cabinet formation under Yeltsin throughout much of the 1990s, but the need to recruit technical expertise into the cabinet became much less powerful once the economy began to grow. [Table 2 about here] Parliamentary fragmentation, presidential legislative support and the size of the communist party Like the economic situation, parliamentary fragmentation and the level of parliamentary support that the presidents enjoyed changed significantly over time. Throughout his period in office Yeltsin never enjoyed a legislative majority, moreover, parliamentary fragmentation was very high until Still, Yeltsin s support varied over time. For a brief period ( ), the most propresidential party was the largest party in the Duma (though it did not control a majority of seats), but from 1995 onward the Communists, a party in outright opposition to the president, became the strongest force in the assembly, controlling a near majority. Parliament s composition, then, presented powerful challenges to Yeltsin s ability to build legislative coalitions. Putin, by contrast, enjoyed majority legislative support for most of his two terms in office. While the Communists remained the largest political force in the assembly after December 1999, the legislature formed a pro-presidential majority coalition part of the way through its term, and the 2003 Duma elections returned an assembly dominated by a single propresidential party, United Russia, with a near two-thirds supermajority (Remington 2005; Remington 2006). Following hypotheses 2 and 3, I expect the president s willingness to recruit ministers with parliamentary and party experience into cabinet to grow with the size of the president s legislative support and the assembly s ability to act collectively. Sub-table 2c shows that while Yeltsin s ministerial choices match these expectations very closely, Putin s ministerial choices diverge dramatically from them. As the upper part 19

20 of the table illustrates, in the Yeltsin governments on average a quarter of the ministers had parliamentary experience, twenty per cent had been affiliated with a party these ministers split evenly between pro-presidential and opposition parties, reflecting the fact that Yeltsin lacked a governing assembly majority throughout his period in office. As the middle and lower parts of table show Yeltsin s appointment strategy responded strongly to changes in the willingness and ability of the assembly to co-operate with him. When a pro-presidential party (Russia s Choice), was the largest party in the assembly (though it lacked a majority), he consolidated the parliamentary links of the government, ensured that a large majority of ministers (67 per cent) had parliamentary experience, and used ministerial appointments to firm up cabinet links with both presidential parties and the opposition (37 per cent had been affiliated with a pro-presidential party, 11 per cent with an opposition party). However, cabinet-assembly ties became much weaker under Yeltsin once the Communists began to dominate the assembly. By contrast, Putin s governing strategy does not appear to have been predicated on links with the assembly. Despite the existence of a pro-presidential coalition, and then a single party majority, only a small minority of the ministers he appointed (14 per cent) had links with pro-presidential parties, an even smaller minority had parliamentary experience (5 per cent), and none had any links with opposition parties (upper part of sub-table 2c). These appointment patterns were invariant to whether the Communists or United Russia was the largest assembly party (lower sections of subtable 2b). In sum although both presidents appointed fully non-partisan cabinets, there were sizable and statistically significant differences in the extent to which they were prepared to use cabinet appointments to consolidate links with the assembly and with legislative parties. The picture that emerges so far regarding the two presidents responses to the structural context suggests differences in the governing strategies of Yeltsin and Putin: Yeltsin included only a small share of ministers in his cabinet to whom he had loyalty links, and sought to consolidate contacts with parliament and parties when that was politically feasible even against the background of an economic crisis, which gave him powerful incentives to prioritize the recruitment of technical experts. Putin, by contrast, chose to prioritize personal loyalty of cabinet ministers over consolidating 20

21 links with the assembly and parties, against the backdrop of a pro-presidential majority, and despite the fact that the pressure to recruit technical experts into the government had subsided with economic growth. The framework outlined above suggests two possible explanations for Putin s ministerial choices which focus on his governing strategy: one possibility is that he inclined toward authoritarian methods of governing; alternatively, his strategy may have been to govern within a democratic framework but through executive prerogative. While I have so far been able examine the impact of observable contextual factors on cabinet composition, I cannot observe a president s governing strategy. In what follows, I therefore proceeded in reverse order, and apply the expectations developed above to the ministerial choices of the two presidents in order to draw inferences about their likely governing strategies. Whether these inferences (and thus the expectations outlined above) are plausible is then evaluated with evidence on the effect that the two presidents had on the use of statutes as opposed to executive prerogative, on the quality of elections, and respect for democratic rights and freedoms. I begin by examining how far the cabinet appointment patterns suggest authoritarian or democratic government strategies. Governing strategies: Authoritarian versus democratic Politicians with authoritarian inclinations would not just be expected to curtail the links of their cabinet to other political actors, they should also rely heavily on ministerial appointees who lack autonomy from the state, and boost the repressive capabilities of their cabinets by raising the share of ministers with experience in the military or with law enforcement backgrounds (H6). Sub-table 2d summarizes how the share of ministers with administrative, military and law enforcement expertise evolved in Russia over time. As the top row of the table illustrates, the reliance on administrative expertise remained extremely high in Russia. This is perhaps less surprising during the initial years after the collapse of the Soviet regime, when an average 93 per cent of the ministers that Yeltsin appointed had prior administrative careers as a minister, government civil servant or a civil servant within the presidential administration. It is, however striking that this pattern remained unchanged as the temporal distance to the Soviet years grew. Putin continued to rely very heavily on administrative expertise, on average 94 per cent of the ministers he appointed had prior administrative careers. In 21

22 fact in the Zubkov government, the last that Putin formed at the end of his second term, in September 2007, nearly 16 years after the collapse of the USSR, 100 per cent of the ministers who were appointed had had prior careers in state administration. The second to fourth rows of sub-table 2d focus on the law enforcement and military expertise that the two presidents drafted into the government and show that Yeltsin placed no particular emphasis on expertise in repression. Nor did Putin appear keen to recruit military men and women and law enforcers directly into government. But Putin was decidedly keener than Yeltsin to recruit administrators who knew how to use the military and law enforcement agencies. Just under 20 per cent of Putin s ministers had expertise of working with the military and law enforcement agencies in the government bureaucracy a sizable and statistically significant increase compared to the Yeltsin era. This appears to have been an area in which Putin sought to bolster the capabilities of his governments. In drawing inferences from the evidence presented so far, I must clearly be cautious, given my reliance on descriptive statistics and the difficulties of working with data that records ministers as falling into multiple categories simultaneously. But once the expectations about the effects of contextual factors and governing strategies on ministerial choices are applied to Russia a fairly consistent picture emerges across the full range of dimensions of government composition: While Yeltsin appointed cabinets under economically difficult circumstances, as a minority president, and in the years following the collapse of the USSR, he appears to have sought to build cabinet links with representative institutions and showed no sign of seeking to boost the cabinet s ability to use repression. Putin by contrast, was significantly less constrained by the economic situation and commanded assembly support, but appeared to curtail the influence of the assembly and parties on the government deliberately while prioritizing personal loyalty, seemed to place a positive value on the autonomy that ministers with state administrative backgrounds have from other political institutions and society at large, and shored up government expertise in managing the military and law enforcement agencies. In other words, while Yeltsin s government appointment strategy signalled a broadly democratic approach, Putin s ministerial choices were suggestive of an authoritarian governing strategy. 22