Presidents and Cabinet Payoffs in Coalition Governments

Abstract

As a key consequence of government formation negotiations among executive and legislative actors, portfolio allocation offers a window to understand the impact of constitutional design and presidential prerogatives on cabinet dynamics across democratic regime types. This article uses Shugart and Carey’s emphasis on the implications of regime distinctions and institutional variation in presidential powers for executive-legislative relations as a starting point for an examination of the extent to which presidents influence government formation outcomes in parliamentary and semi-presidential systems. Many presidents in these political systems have some influence on government formation, which enables them to shape cabinet composition. Yet, whether these powers advantage presidential parties in reaping more cabinet spoils than their proportional share has yet to be investigated. Using data on 442 government formation situations in 23 European parliamentary and semi-presidential democracies during 1945–2015, this study finds that parties of strong presidents generally, or presidents formally empowered to choose a formateur, are more likely to be advantaged in the allocation of cabinet seats than their peers who are not allied with the head of state.

Keywords

coalition bargaining cabinet proportionality government formation presidential powers semi-presidentialism

Government formation is one of the most consequential political processes regardless of whether it takes place in a parliamentary, semi-presidential or presidential system of government. Shugart and Carey’s 1992 ground-breaking book, Presidents and Assemblies, highlighted the main institutional distinctions between regime types based on the balance of constitutional powers between executive and legislative branches. However, their analysis paid less attention to cabinet politics and the crucial ways in which cabinets link the executive and legislative branches of government (Samuels, 2009: 710). This contribution to the PSR Special Issue on Shugart and Carey’s 1992 seminal volume focuses on the relationship between presidential powers and outcomes of government formation. Specifically, we ask whether variation in the constitutional powers granted to the head of state and the complexity of the bargaining context in which a new government forms affect the extent to which presidents are able to tilt the balance of cabinet appointments in favour of their parties not only under separation of powers regimes, where they always play the role of formateurs, but also in semi-presidential and parliamentary systems, where their role in government formation has been considerably less studied. This research question is firmly anchored in the agenda set by Shugart and Carey’s pivotal book, as it deals with consequences of institutional variation on government formation outcomes within and among different regime types beyond pure presidential systems.

Although Shugart and Carey stopped short of discussing the implications of institutional variation for cabinet politics, they highlighted the important consequences that the allocation of presidential powers has for regime durability and performance. As heads of state with strong legislative powers have few incentives to negotiate with the legislature, Shugart and Carey argued, presidential regimes are more likely to experience crises and failure. In contrast, presidents whose powers over legislation are constrained have little choice but to regulate conflict through negotiations with the legislature. Subsequently, scholars have focused on variation in the president’s constitutional prerogatives to help explain outcomes of cabinet politics, including the allocation of cabinet portfolios among coalition parties (Amorim and Neto, 2002, 2006; Alemán and Tsebelis, 2011; Amorim Neto and Samuels, 2010; Chaisty et al., 2014). For example, the cross-regional comparison carried out by Chaisty et al. (2018) has shown that the ability to distribute cabinet seats to reward loyal coalition partners is the presidents’ most important tool of coalition management, confirming these actors’ strong motivation to influence the composition and the division of cabinet spoils.

A similar logic to coalition-building underpins non-presidential systems. In contrast to presidential regimes, the presidential party may not always be in government in multi-party parliamentary and semi-presidential systems, where the identity of the formateur and the composition of the cabinet depend on inter-party negotiations over executive coalitions. Although most parliamentary and semi-presidential constitutions, explicitly or implicitly, assign at least some discretion to the head of state in either directly appointing the prime minister, or designating the individual in charge of conducting negotiations to form a government (the formateur), few studies have incorporated presidents in models of parliamentary bargaining over executive coalitions. Indeed, the coalition theory literature has traditionally assumed that heads of state in parliamentary systems are non-strategic in appointing a formateur (Bassi, 2013) and have no preferences over potential governments (Morelli, 1999). Nevertheless, heads of state in non-presidential systems have strong incentives to deploy their powers to advantage their own party in the division of cabinet spoils. For example, being able to influence this process may enable presidents to hand-pick individual ministers (Protsyk, 2005; Tavits, 2008). Moreover, the inclusion of the presidential party in government may influence the policymaking process to the president’s advantage. By aligning the cabinet and the president’s policy goals, the potential for conflict and delays in policymaking is reduced and presidents are in a better position to advance their policy agenda (Urbanavicius, 1999).

Indeed, numerous studies have drawn attention to the significant variation in the role that heads of state in non-presidential systems play in government formation, ranging from an active involvement in selecting the prime minister (PM) and cabinet to merely formally appointing a PM who was nominated and selected exclusively by parliament (Bäck and Dumont, 2008; Bucur and Cheibub, 2017; Laver and Schofield, 1990). For example, Strøm and colleagues (1994) counted heads of state among the coalition bargaining constraints that may reduce the range of coalition options. Importantly, these authors credited the head of state with an agenda-setting role in government formation and expected the president’s party to gain bargaining power under certain circumstances (Strøm et al., 1994: 320). More recent empirical studies pointed out a strong correlation between presidential powers and the share of non-partisan ministers (Amorim Neto and Strøm, 2006; Tavits, 2008) and argued that partisan presidents have considerable leeway in choosing formateurs and deciding their exploratory scope (Bucur, 2017; Carroll and Cox, 2012). Moreover, speaking about cabinet politics as a ‘missing link’ in the study of the separation of powers, Samuels (2009: 712) argued that cabinet dynamics in semi-presidential systems represent a middle ground between parliamentarism and pure presidentialism. Consequently, Samuels argued, parties may expect lower office and/or policy payoffs in semi-presidential systems relative to parliamentary systems (Samuels, 2009: 710–711). However, while several studies have compared the proportionality of the distribution of cabinet portfolios as a function of regime attributes (Amorim Neto and Samuels, 2010; Samuels, 2009), they stopped short of investigating whether presidential parties in parliamentary and semi-presidential systems do exploit this institutional source of bargaining power to reap more than their fair share of cabinet portfolios. This article addresses this gap and asks whether presidents with more extensive prerogatives in government formation are more effective in generating portfolio bonuses for their political allies across non-presidential democracies.

In sum, the literature on government formation has produced considerable evidence that presidents influence various aspects of coalition outcomes, including the duration of bargaining processes (Savage, 2018), selection of PMs (Bucur and Cheibub, 2017; Glasgow et al., 2011; Kang, 2009), the share of non-party ministers in cabinet (Amorim Neto and Strøm, 2006; Schleiter and Morgan-Jones, 2009; Tavits, 2008), and the choice of specific coalitions (Kang, 2009). To identify the mechanisms through which the president’s influence on government formation is realised, scholars have focused on various constitutional factors such as direct and indirect elections (Bucur and Cheibub, 2017; Glasgow et al., 2011; Kang, 2009), the vote of investiture (Glasgow et al., 2011), the power to nominate the formateur (Bucur and Cheibub, 2017) and a combination of legislative and non-legislative powers (Savage, 2018). In terms of cabinet composition, alliance with the president has been singled out as a determinant of formateur advantage in portfolio allocation in parliamentary and semi-presidential systems. Carroll and Cox (2012) argued that formateurs allied with presidents have a larger exploratory mandate, which strengthens their bargaining position and ability to garner a portfolio bonus. Thus, the idea that the president’s appointment powers matter for the degree of cabinet proportionality lends itself to comparisons across and within regime types. However, to date empirical studies on the extent to which heads of state in non-presidential systems are able to skew the distribution of cabinet payoffs in favour of their parties are still missing. This is the aim of this article, which asks whether differences in the constitutional powers granted to presidents in parliamentary and semi-presidential systems and the bargaining complexity of government formation processes in these regime types affect the ability of presidents to influence their party’s share of cabinet spoils.

Presidential Parties and Portfolio Allocation

Shugart and Carey’s (1992) emphasis on variation in the constitutional roles assigned to presidents and their distinction between president-parliamentary and premier-presidential presidential sub-types inspired scholars to investigate the extent to which variation in the presidents’ formal powers make a difference for their ability to shape cabinet composition (see Elgie, 2016 for a comprehensive review). However, while focusing on the relationship between legislatures and executives, Shugart and Carey did not explore the extent to which the relationship between presidents and their parties changes under different constitutional frameworks and how it impacts political outcomes. This important question was addressed by Samuels and Shugart (2010), who argued that the separation of powers has important consequences for party organisation and behaviour. Their main claim was that the more the origin and survival of the executive and legislative branches are independent of the other, the more parties will tend to presidentialise (Samuels and Shugart, 2010: 38). Here, ‘party presidentialisation’ refers to the separation of parties into two distinct branches, one in the legislature and the other in the executive, which pursue different strategies for candidate selection, campaigning and governing. For example, Samuels and Shugart argue, the presidentialisation of parties is noticeable when political outsiders are more likely to hold office or when party influence over prime-ministerial dismissal decreases.

A key implication of the ‘presidentialisation of parties’ thesis is that semi-presidentialism, where the president is elected separately from the legislature and the government may be accountable to both the president and the legislative, is more presidential than parliamentary. Thus, Samuels and Shugart argued, the existence of a directly elected president tends to ‘contaminate parties in semi-presidential systems, interfering in the principal-agent relationship between parties and their prime-ministerial agents in the legislature’ (Samuels and Shugart, 2010: 255). Examples of this phenomenon can be found in the powerful role presidents play in the selection and dismissal of the PM, even in premier-presidential systems, where the cabinet is formally accountable only to the legislature. Consequently, the presidents’ ‘contamination’ of the party-PM relationship is bound to reduce parties’ control over the composition and direction of government (Samuels and Shugart, 2010: 256). Therefore, the presidentialisation of parties may affect cabinet politics as well.

The proportionality of cabinets is an important observable implication of the impact of separation of powers on cabinet politics. Unlike parliamentary cabinets, where PMs must make sure that seats are proportionally distributed to coalition partners to ensure their survival (Amorim Neto, 2006: 436), whether or not cabinets are proportional in presidential systems depends on the presidents’ governing strategies. If they choose to rely on statutes as policymaking strategies, then the proportionality of cabinets will reflect their legislative coalition; conversely, if presidents decide to pursue their policy goals through decrees or other unilateral powers, the incentives for coalition-making and cabinet proportionality decrease (Samuels, 2009: 711–712). To the extent that parties presidentialise under semi-presidential regimes and see their influence on government composition reduced, the distribution of cabinet portfolios may also deviate from proportional outcomes. Thus, Samuels and Shugart (2010: 52) anticipate that strong presidents in semi-presidential systems should be less committed to the principles of full partisanship and proportional distribution of portfolios across coalition partners than PMs in pure parliamentary systems. Similarly, Samuels (2009) argues that the expected payoff for parties in terms of ‘office’ and/or ‘policy’ benefits should be lower in semi- and pure presidential systems relative to parliamentary systems” (Samuels, 2009: 711).

Several studies have confirmed that cabinet dynamics in semi-presidential systems represent a ‘middle ground’ between parliamentarism and presidentialism as far as the proportional distribution of cabinet portfolios to coalition partners is concerned (Amorim Neto and Samuels, 2010; Samuels, 2009). However, while these studies have compared the proportionality of cabinet allocations across regimes types, they have not looked at whether the president’s party is advantaged in portfolio allocation. For example, we do not know whether deviations from proportionality in semi-presidential systems reflect a strengthening of the small party bias typical for parliamentary cabinets (Warwick and Druckman, 2001, 2006) or are due to strong presidents who are able to shape coalition formation and carve a higher portfolio share for their party. Furthermore, while in presidential systems the president is always the formateur, in semi-presidential systems the president’s party plays the formateur role only in unified executive situations, when both the president and the PM belong to the same party. However, the president’s party may take part in coalitions without holding the PM post, as is the case in divided executive situations when the president and PM belong to different parties and the president’s party plays the role of a junior coalition partner. Thus, presidential parties may benefit from the president’s partisan involvement in coalition formation either from formateur or non-formateur positions. Therefore, if we want to investigate the extent to which presidents influence cabinet proportionality in non-presidential systems, then we need to take into account the office payoffs derived by all types of presidential parties, be they formateur or non-formateur parties.

One difficulty with capturing presidential influence on cabinet politics in semi-presidential systems arises from the wide variation in presidential powers. Previous works have shown that the mode of presidential election is not a good predictor for the president’s influence on political outcomes (Elgie, 2015; Tavits, 2008). The direct election of the president does not necessarily mean that the president is a powerful political actor. Indeed, many directly elected presidents are weak, either due to the restricted number of constitutional powers vested in the office or because the historical conditions under which presidencies were established. Conversely, even if parliament-selected presidents operate within a system in which the PM is the dominant actor within the executive branch, some presidents hold important powers that allow them to emerge as significant political actors at crucial moments (Elgie, 2015: 322–323). Therefore, more fine-grained indicators than the mode of presidential election are needed to capture presidential practice. As Elgie and Passarelli (2018: 4) suggest, it is plausible to expect an association between higher levels of presidential power and party presidentialisation. Therefore, instead of linking regime type and presidential practice to cabinet proportionality, one can hypothesise that if presidents use their influence on cabinet formation in a partisan fashion to benefit their political allies, then they are more likely to do so when they have strong constitutional powers. Devising an empirical test on this basis has the advantage of generating variation in parliamentary countries as well, allowing us to include parliamentary countries in the test and carry out a full comparison of presidential influence on portfolio allocation across all regime types. These expectations yield the following hypothesis about presidential power and portfolio payoffs for presidential parties:

Presidential Power Hypothesis: Controlling for their size, presidential parties are more likely to receive a bonus when it comes to portfolio allocation at higher levels of presidential power.

Theories that incorporate the role of presidents in coalition formation must also consider the circumstances that allow the effective use of cabinet appointment powers. For example, Bucur and Cheibub (2017) argued that presidents, however they are elected, can exploit a constitutionally proactive role in government formation to favour their own parties for the PM position only if no clear-cut winner emerges from a general election. Thus, whenever the distribution of power in parliament clearly favours one party, the PM post is typically occupied by the largest legislative party. Under conditions of high bargaining uncertainty, however, directly elected presidents who are constitutionally enabled to participate in the formation process do act to increase the chance that a co-partisan will become PM.

The way in which political uncertainty and a fragmented party system affect the outcome of negotiations over cabinet formation and portfolio allocation has been considered in the coalition formation literature. For example, Baron and Ferejohn’s (1989) bargaining model predicts that the bonus extracted by the formateur due to its agenda-setting power increases as the party system becomes more fragmented. Empirically, however, Falcó-Gimeno and Indridason (2013) find that portfolio allocation is closer to proportionality in uncertain and complex negotiation processes. According to this theory, forming a government coalition is a more challenging task in difficult bargaining situations and negotiating partners resort to a more proportional allocation to solve bargaining problems. Extending this idea, Carroll and Cox (2012) propose that formateurs may reap a portfolio bonus in complex bargaining situations if they are allied with the president. Accordingly, some presidents may affect the formateur’s bargaining position by either explicitly limiting their coalition options or giving out an ‘exploratory mandate’ that authorises them to consider multiple possible coalitions. Thus, Carroll and Cox argue, the more exploratory the mandate granted by the president, the stronger the bargaining position of formateurs and their ability to derive a portfolio bonus. From this point of view, the value of having a presidential ally increases as the party systems becomes more fragmented, given that the president has little discretion to choose a formateur or grant a larger exploratory score in less fragmented cases (Carroll and Cox, 2012: 10–13). These expectations are corroborated by findings from Schleiter and Morgan-Jones (2010), who found that as the fragmentation of party groups in parliament increased, so did the president’s influence over formation outcomes. This idea leads to the following extension of the presidential power hypothesis:

Bargaining Environment Hypothesis: The greater the president’s constitutional powers and the fragmentation of the legislative party system, the more likely presidential parties are to obtain a greater share of cabinet portfolios than parties of a similar size.

Data and Variables

The expectations outlined above are investigated using a dataset that consists of portfolio allocations to 1451 parties that participated in 442 coalition governments in 23 European democracies (plus Israel and Turkey) with parliament-selected or directly elected heads of state from 1945 to 2014¹. Distinctions are made between parliamentary and semi-presidential systems broadly defined, without differentiating between the premier-presidential and president-parliamentary sub-types identified by Shugart and Carey (1992). Undoubtedly, the inclusion of president-parliamentary regimes, which concentrate a high degree of formal power in the presidential office, would enable a more comprehensive test of the link between variation in presidential prerogatives and cabinet dynamics. Unfortunately, though, the sources of government formation data noted below do not include former Soviet Union states, where most president-parliamentary regimes are located. Although three countries in the current sample are formally classified as president-parliamentary regimes, namely Austria, Iceland and Portugal (1976–1982), executive politics in these countries are seen as being fully parliamentarised (Müller, 1999; Shugart, 2005). As a result, differentiating among semi-presidential sub-types in this sample is unlikely to enable meaningful conclusions about inter-regime variation and portfolio allocation outcomes. Instead, different measures of presidential powers are used to capture variation in formal prerogatives that are considered essential for studying presidential influence on political outcomes across distinct regime types.

A new government is counted when there is a change in the party composition of a government coalition; a change in the identity of the PM; or a general parliamentary election. Caretaker and technocratic cabinets, as well as situations where a single party controlled the majority in parliament, are excluded, as the article’s focus is on the formation of fully empowered, party-based coalition governments (further information about the number of coalition governments and time period under study in each country is available in the Online Appendix, Table A1.1).

The unit of analysis is a political party in a government coalition and the dependent variable records the share of ministries allocated to a governing party. Raw seat shares, which have been empirically proven to predict the distribution of cabinet portfolios in European governments better than theoretical voting weights (Cutler et al., 2016), are used to measure each party’s Seats contribution to governing coalitions. The data on legislative seat shares and portfolio allocation are based on the ParlGov database (Döring and Manow, 2016), Woldendorp et al. (2000), Seki and Williams (2014) and Williams and Seki (2016). These data were updated and crosschecked against the information published in the annual country reports published in the European Journal of Political Research Political Data Yearbook since early 1990s until 2014. Where available, and especially when the above-mentioned sources provided different information, official data sources to crosscheck, update and correct existing portfolio allocation data were used (general and country-specific sources are provided in the Online Appendix, Table A1.2). Following the existing literature, the posts held by non-partisan ministers are not included in the analysis to keep focus on partisan dynamics. The coding of the presidents’ political affiliation is based on ParlGov and the World Political Leaders database (Zarate’s Political Collections, 2015).

A number of independent variables are operationalised to test the two hypotheses outlined above. Semi-presidentialism differentiates between systems with parliament-selected and popularly elected presidents – that is, between parliamentary and semi-presidential systems. Formateur status equals 1 for PM parties and 0 otherwise. Presidential Party indicates the president’s party. Two measures are used to explore the presidential power hypothesis. The first measure captures the effect of presidential power generally and is based on the Presidential Power Index developed by Doyle and Elgie (2016)². The second measure focuses on the president’s influence on cabinet appointments and indicates whether presidents play a proactive or passive role in the appointment of the PM (based on Bucur and Cheibub, 2017). The coding of the Proactive Nomination variable is based on constitutional texts. Proactive presidents are recorded if the constitution grants them the unconstrained power to nominate a PM. For example, this is the case in the Fifth French Republic, where the constitution stipulates that ‘the president appoints the prime minister’ (Article 8). Passive presidents are recorded when their participation is limited to appointing a head of government nominated and chosen by the parliament (like in Ireland and Germany); or to nominating a premier designated by the constitution (like in Bulgaria, where the president is required to nominate the leader of the largest party). Table 1 groups cases in the dataset based on regime types and the presidents’ constitutional role in government formation. The distribution of cases is shown in Table 2. This classification shows that the president’s role in government formation cuts across the two regime types. Thus, it is not the case that directly elected presidents in semi-presidential systems are more likely to have a constitutionally defined active role in government formation than those who are indirectly elected in pure parliamentary systems. Finally, the Effective Number of Parliamentary Parties (ENPP) (Laakso and Taagepera, 1979) is used as a measure of bargaining complexity to test the second hypothesis about the joint effect of presidential power and legislative fragmentation. Descriptive statistics and information about data sources and the coding of all variables are available in the Online Appendix, Tables A1.3–A1.5.

Table 1.

Heads of State in Non-presidential Systems and Roles in Government Formation.

		President’s role in government formation
Regime type		Proactive	Reactive
	Semi-presidential	Austria (1951–2014)	Bulgaria (1992–2014)
		Czechia (2013–2014)	Croatia (2000–2014)
		France (1965–2014)	Finland (2000–2014)
		Iceland (1945–2014)	Ireland (1945–2014)
		Finland (1945–1999)	Poland (1993–2014)
		Poland (1991–1992)	Slovenia (1992–2014)
		Portugal (1976–2014)
		Romania (1991–2014)
		Slovakia (1999–2014)
		Turkey (2014)
	Parliamentary	Austria (1945–1950)	Bulgaria (1991)
		Czechia (1993–2012)	Estonia (1992–2014)
		France (1946–1964)	Germany (1949–2014)
		Hungary (1991–2014)	Greece (1975–2014)
		Italy (1948–2014)	Israel (1968–2014)
		Israel (1949–1967)
		Latvia (1993–2014)
		Poland (1989)
		Slovakia (1993–1998)
		Turkey (1983–2013)

Table 2.

Heads of State in Non-presidential Systems and Roles in Government Formation (Distribution of Cases).

		President’s role in government formation
		Proactive	Reactive
Regime type	Semi-presidential	159	61
Regime type	Parliamentary	138	84

Analysis and Results

Before assessing the extent to which the office spoils reaped by presidential parties vary under different circumstances, general patterns of portfolio allocation are evaluated. Figure 1 plots seat share contributions to coalition governments against cabinet portfolio shares and presents the fit of this relationship (i.e. ordinary least squares (OLS) regression lines) for formateur and non-formateur parties in parliamentary (Panel 1) and semi-presidential systems (Panel 2). As is apparent from the figure, although cabinet portfolios appear roughly proportionally allocated across the two regime types, large parties tend to be under-compensated in the average portfolio allocation. Exceptions to this trend exist, however, as a non-negligible number of formateur parties appear over-compensated in both regime types (being situated above the 45° black line that indicates one-to-one proportionality).

Figure 1.

Portfolio Share and Seat Contribution Across Parliamentary and Semi-presidential Systems.

A closer look at patterns of allocations to both formateur and non-formateur parties reveals some differences across the two regime types. In parliamentary systems, the changeover from over-compensation to under-compensation (i.e. the point where the regression line crosses the one-to-one proportionality line) occurs at a seat share of 46.7% for formateur parties. This trend confirms the under-payment effect of formateur parties found by previous studies (Warwick and Druckman, 2006: 651), as the average formateur party contributes more than this percentage of their coalition’s legislative weight. Indeed, the average formateur in parliamentary systems provides 58.7% of the government’s total legislative weight but receives just 55.8% of the portfolio payoff. By comparison, the average non-formateur party provides 16.3% of the cabinet’s seat share and receives 17.4% of its payoff. This gap increases in semi-presidential systems, where the changeover from over-compensation to under-compensation occurs at a seat share of 25.1% for formateur parties and 41.4% for the other parties. In these systems, the formateur party provides 60.8% of the government’s total legislative weight, but receives just 57.9% of the portfolio payoff, while the average non-formateur party provides 20.7% of the cabinet’s seat share and receives 22.1% of its payoff. Overall, though, patterns of portfolio allocation are roughly similar in both regime types, corroborating the slight formateur underpayment and small party bonus.

The next sets of analyses investigate whether being the president’s party makes a difference for the share of cabinet payoffs reaped by formateur and non-formateur parties under different institutional and political contexts. The multivariate analysis is based on OLS regressions.³ The results are presented as a set of marginal effect plots, while the tables laying out the regression models are relegated to the Online Appendix.

We start with a basic model that regresses the cabinet portfolio share of each party on its legislative seat share contribution to the governing coalition. To test whether the effect of formateur status varies with party size and regime type, the party’s status as formateur and the regime type are interacted with its legislative seat share (results shown in the Online Appendix, Table A2.1, Model 1). Panel 1 in Figure 2 shows the marginal effect of formateur status across party size in parliamentary and semi-presidential regimes, which represents the fitted formateur advantage at a range of seat contributions to the coalition⁴. 95% confidence intervals around the sloping lines determine the conditions under which the formateur status has a statistically significant effect on the size of cabinet payoffs. A histogram portraying the frequency distribution of cabinet parties in parliamentary and semi-presidential systems across the seat share range is superimposed over the marginal effects plot. The scale for this distribution is given by the vertical axis on the right-hand side of the graph.

Figure 2.

Marginal Effects of Formateur and Presidential Party Status on Portfolio Allocation, Parliamentary and Semi-presidential Systems Compared. (1) Formateur. (2) Presidential Party.

Panel 1 in Figure 2 shows a positive but declining effect of formateur status on portfolio allocation in parliamentary systems. The positive effect is significant⁵ for parties that contribute up to 64% of the coalition’s legislative seat shares. The histogram shows that 86.3% of the observations in the parliamentary sample fall into this range of statistical significance. This result confirms the underpayment of large formateur parties reported by previous studies (Warwick and Druckman, 2001, 2006), indicating that junior partners are over-compensated in these cases. In other words, formateur parties may receive more compensation than other parties, but only when their seat share is relatively small. These results also show that the formateur party status does not play a significant role in semi-presidential systems.

To investigate whether the presidential party status might be playing a significant role in these systems, we re-estimate Model 1 by interacting the seat share and regime type variables with the presidential party indicator (results shown in the Online Appendix, Table A2.1, Model 2). The marginal effect of presidential party status on portfolio allocation is presented in the second panel in Figure 2. This time around we can see a significant presidential party effect in semi-presidential systems, which generates a portfolio bonus for these parties that barely varies with party size. The Panel 2 results indicate that being the president’s party matters in parliamentary systems as well. In this case, the positive effect is declining and significant for parties that contribute less than 52% of the coalition seats, which covers 82% of the observations.

Do formateur and non-formateur parties equally benefit from being the president’s party? To answer this question, we interact both Presidential Party and Formateur indicators with seat shares and the regime type (shown in the Online Appendix, Table A2.1, Model 3). Figure 3 presents the marginal effect of presidential party status across seat contributions for formateur (Panel 1) and non-formateur parties (Panel 2). Both panels present the effects corresponding to parliamentary and semi-presidential systems separately.

Figure 3.

Marginal Effects of Presidential Party Status on Portfolio Allocation for Formateur and Non-formateur Parties by Seat Share, Parliamentary and Semi-presidential Systems Compared. (1) Formateur. (2) Non-formateur.

The results presented in Figure 3 show different effects of presidential party status for formateur and non-formateur parties. Panel 1 indicates that the presidential party status benefits formateur parties only in semi-presidential systems. Moreover, it is large formateur parties that seem to benefit from being the president’s party – as the effect is positive and statistically significant for formateur parties that contribute more than 54% of the coalition seats (covering 61% of the sample, according to the superimposed histogram). Conversely, according to Panel 2, most non-formateur parties benefit from being the president’s party in both systems, regardless of their legislative size.⁶ While the effect on non-formateur parties may be confounded by their size, given the apparent bias in favour of small parties highlighted by previous works (see, for example, Warwick and Druckman, 2001, 2006), the fact that the presidential party status benefits large formateurs is at odds with previous findings of formateur under-compensation.

The next sets of models examine whether the presidential party effect is conditioned by formal presidential powers. Two measures are used to test the President Power Hypothesis. First, the effect of presidential power generally is estimated using the index developed by Doyle and Elgie (2016). For this analysis, we examine a model where presidential power scores are interacted with seat shares and the dummy variables for formateur status, presidential party and regime type⁷ (results shown in the Online Appendix, Table A2.2, Model 1). Presidential power scores at the 75th and 25th percentiles in the data are used to compare effects at strong and low levels of presidential power⁸. The marginal effects of presidential party status for formateur and non-formateur parties depending on whether the presidency is strong or weak are presented in Figure 4. To illustrate the frequency of observations at high and low ends of presidential power, the histograms superimposed on these plots are limited to observations from government formations where presidential power scores are above or below these thresholds.

Figure 4.

Marginal Effect of Presidential Party on Portfolio Allocation for Formateur and Non-formateur Parties by Presidential Power, Parliamentary and Semi-presidential Systems Compared. (1) Total Power-Strong President. (2) Total Power-Weak President.

First, we focus on cases with a strong presidency. The results in Panel 1 indicate that holding the country’s presidency, even when it is generally strong, has no impact on formateur parties in parliamentary systems. The effect on non-formateur parties is positive and significant for most of the observations in the sample but shows little variation with size⁹. Conversely, in semi-presidential systems a substantial number of formateurs benefit from being allied with a strong president. Specifically, we see a positive and significant effect on medium and large formateurs that contribute more than 58% of the coalition seats, which make up 52.4% of the sample. Substantially, this effect translates into a portfolio bonus of 5%–13% for the president’s party. Non-formateurs seem to benefit from the alliance with a strong president as well. The effect is positive and significant for nearly all observations in the sample and translates into a portfolio bonus ranging between 6% and 18%. While this result seems to confirm our expectations, we need to test whether different patterns occur when the presidency is weak. Results in Panel 2 indicate that formateurs in parliamentary systems, who do not benefit from being allied with strong presidents, are not harmed by an alliance with weak presidents either. The effect on non-formateurs is largely insignificant as well. As far as semi-presidential systems are concerned, while we can still see that being the president’s party has a positive effect on medium and large formateurs, this time around the effect size is much smaller (i.e. around 4%–5% for all parties). Similarly, a weak presidency still seems to generate a 6%–7% bonus for non-formateur parties regardless of their size.

Next, we limit this analysis to presidential power in government formation. To estimate the effect of this specific element we re-estimate the previous model using a dummy variable that captures the president’s constitutionally unconstrained power to nominate the PM. These results are shown in the Online Appendix (Table A2.2, Model 2), while marginal effects plots are presented in Figure 5.

Figure 5.

Marginal Effect of Presidential Party on Portfolio Allocation for Formateur and Non-formateur Parties by Presidential Power, Parliamentary and Semi-presidential Systems Compared. (1) Appointment Power-Strong President. (2) Appointment Power-weak President.

As we can see in Panel 1, in parliamentary systems formateur parties are still not affected by an alliance with strong presidents. Most non-formateurs, on the other hand, do seem to extract a bonus ranging between 2.8% and 11.5% as their size increases. These effects are much stronger in semi-presidential systems, where the majority of medium and large formateur parties that contribute more than 52% of the coalition seats (i.e. 60% of the sample) are likely to reap a bonus between 3% and 14% of the cabinet seats. Similar bonuses (i.e. 3%–17%) are likely for nearly all non-formateur parties in the semi-presidential sample.¹⁰ While these results are similar to those corresponding to the general measure of presidential power, comparisons with cases of weak presidencies present starker differences. As the results shown in Panel 2 indicate, when presidents lack discretion in the nomination of the formateur, we no longer observe a bonus for formateurs and non-formateurs in either regime type. Thus, using a measure that taps more directly into presidential influence on cabinet composition provides stronger evidence for our expectation that parties allied with strong presidents in semi-presidentialism might take advantage of an important source of bargaining power compared to their peers in semi-presidential systems with a weak presidency and in pure parliamentary systems. This might explain portfolio allocation patterns in the Fifth French Republic, where the head of state has considerable influence on government formation and formateur parties enjoy more advantages in portfolio allocation than in any other country in Western Europe (Bucur, 2018). Even when the president’s party participated in government as a junior coalition party, it still managed to reap a portfolio bonus due to the president’s intervention. In 1974, President Giscard d’Estaing of the centrist Independent Republicans rewarded Jacques Chirac of the Gaullist party with the premiership in 1974 for having supported his candidacy at the first ballot at the expense of the Gaullist candidate Chaban-Delmas. However, in the ensuing coalition government where the Gaullist party contributed nearly 61% of the seats and the president’s party just 18%, the PM’s party secured only one portfolio more than the president’s party. This allocation resulted in the formateur party’s under-payment of about three cabinet seats, while the president’s party gained two more seats than its proportional share. This example illustrates that a strong presidency might prove an important source of bargaining power in semi-presidential systems even when the president’s party is the junior party in coalition.

Next, we turn to the Bargaining Environment Hypothesis, which suggests that the presidential party is more likely to be over-compensated in portfolio allocation as the president’s constitutional powers and the fragmentation of the legislative party system increase. This hypothesis is based on the expectation that when alternative coalitions can be formed, the constitutional setting allows the president to become the architect of coalition formation. Thus, the final model specification includes the ENPP in interaction with the variables denoting the regime type, the formateur, the president’s party, presidential power and seat share contribution. As before, we conduct this analysis using both the general measure of presidential power and the president’s discretion in the nomination of the formateur (results shown in the Online Appendix, Table A2.3, Model 1). The marginal effects of presidential party status for formateur and non-formateur parties depending on whether the presidency is generally strong or weak and shown over the observed range of effective parliamentary parties are presented in Figure 6.

Figure 6.

Marginal Effect of Presidential Party on Portfolio Allocation for Formateur and Non-formateur Parties by Presidential Power and Legislative Fragmentation, Parliamentary and Semi-presidential Systems Compared. (1) Total Power-Strong President. (2) Total Power-weak President.

In line with the previous analysis, the results in Panel 1 indicate that holding the country’s presidency, even when it is generally strong, has no impact on formateur parties in parliamentary systems regardless of legislative fragmentation. The effect on non-formateur parties is positive and significant when ENPP exceeds 3.25, which covers 85% of the parliamentary sample. The substantial effect is relatively modest at about 3%–4% and varies little across the observed values of ENPP. By comparison, in semi-presidential systems a substantial number of formateurs benefit from being allied with a strong president. Specifically, we see a positive and significant effect ranging between 6% and 14% when ENPP exceeds four legislative parties. This expected increase amounts to 71% of the range of parties in the semi-presidential sample and, therefore, indicates a substantively important estimated effect. Junior coalition parties benefit from the alliance with a strong president as well. We can see a positive and significant effect ranging between 27% and 3% (i.e. higher bonuses extracted for lower values of ENPP) when the legislative party system is limited to about five effective parties, which is the case in about 63% of the observations.¹¹

These effects are considerably smaller when the presidency is weak. The results in Panel 2 confirm that alliance with a weak president does not benefit formateurs. However, we can still see a significant effect for about 60% of the junior coalition parties for ENPP values below 5. Similarly, while the positive effect for formateur parties disappears in semi-presidential systems as well, junior coalition parties seem to enjoy the same advantages as their peers in parliamentary systems. Thus, the lack of variation in the effect of presidential party status on portfolio allocation for non-formateur parties at high and low levels of presidential power suggests that the portfolio bonus is more likely to be explained by the known small party bias effect than by the alliance with the president. However, the large variation in the substantive magnitude and statistical significance of the effect of presidential party status on formateur parties in semi-presidential systems at high and low levels of presidential power lends support to the second hypothesis as far as formateur parties are concerned.

These results are robust to using the presidential power measure that is limited to nomination prerogatives. To estimate the effect of this specific power we re-estimate the previous model using a dummy variable that captures the president’s constitutionally unconstrained power to nominate the PM. These results are shown in the Online Appendix (Table A2.3, Model 2), while marginal effects plots are presented in Figure 7.

Figure 7.

Marginal Effect of Presidential Party on Portfolio Allocation for Formateur and Non-formateur Parties by Presidential Power and Legislative Fragmentation, Parliamentary and Semi-presidential Systems Compared. (1) Appointment Power-Strong President. (2) Appointment Power-weak President.

The results in Panel 1 show that in parliamentary systems formateur parties allied with a strong president obtain a portfolio bonus of about 4.8%–15.6% (i.e. the portfolio bonus increases with fragmentation, as suggested by the ascending regression line) under conditions of high legislative fragmentation (i.e. for ENPP values higher than 5.25). About 29% of the formateur parties in the sample fall into this region of significance. A rather uniform positive effect (i.e. not varying much as a function of fragmentation) ranging from 3% to 5.2% also applies to almost all junior coalition partners (i.e. 96% of the observations in the sample, across ENPP values above 2.75).

These effects are even stronger in semi-presidential systems. Formateur parties allied with a strong president obtain a portfolio bonus of about 6%–4.7% (note that this time around the portfolio bonus slightly declines with fragmentation) in a range between 2.75 and 5 legislative parties. This is a substantially important effect as it covers 58.5% of the formateur parties in the semi-presidential sample. We should also note that nearly 80% of the observations in the semi-presidential sample are situated at ENPP values equal or below 5. For example, if we lowered the significance level at 10%, then the positive effect of presidential party status would cover 83% of the sample. Somewhat higher bonuses (i.e. 17.4%–7.4%) are likely for non-formateur parties for ENPP values that are lower than 5.5, which includes 84.7% of the observations.

In contrast, as shown in Panel 2, when presidents lack discretion in the nomination of the formateur, we no longer observe a bonus for formateur parties in either regime type. Moreover, this time around, only one-quarter of the non-formateur parties are advantaged in portfolio allocation under conditions of small and medium legislative fragmentation (i.e. for ENPP values smaller than 3.5). The effect appears stronger in semi-presidential systems, where non-formateur parties allied with a weak president still get a bonus of 10%–1% (the effect size decreases as ENPP increases) when there are up to 4.5 legislative parties. About 38.5% of the junior coalition parties fall in this area of significance. However, we can also observe a negative effect of presidential party status of about 1.6%–11.8% under conditions of high legislative fragmentation (i.e. when ENPP is higher than 4.5). This finding applies to 24% of the range of junior coalition partners in the semi-presidential sample and, therefore, indicates a non-negligible estimated effect.

Overall, the second hypothesis receives satisfactory support in both parliamentary and semi-presidential systems for formateur parties, as the effect of the presidential party status is significant in complex bargaining environments only when the president has discretion in choosing the formateur. Thus, under conditions of high bargaining complexity, alliance with a strong president can modify the negative effect of formateur status on large parties found by previous comparative studies (Browne and Franklin, 1973; Browne and Frendreis, 1980; Warwick and Druckman, 2001, 2006). An example in this regard could be Finland before the revision of the constitution in 1999 reduced the president’s powers. Arter (1999: 65) notes that ‘presidential influence in domestic matters has been most evident in the sphere of coalition-building, aided in some measure by the polarized nature of the Finnish party system and its concomitant government instability’. Thus, between 1966 and 1987, President Kekkonen used his discretion in government formation to ‘harness the three main parties of the centre-left (Centre, Social Democrats, and Communists) in a relatively durable presidential majority while excluding the National Coalition (Conservative) Party from government despite significant gains at the polls’ (Arter, 1999: 55, 58). Following the constitutional revision of 1999 that stripped the presidency from most of its powers, however, the head of state no longer plays an active role in bargaining over coalition formation.

Romania provides another example for situations in which the President may be able to cause parties to realign in view of government formation when the election outcome is less clear (see Bucur and Cheibub, 2017: 806). After the concurrent presidential and legislative elections held in November 2004, the left-wing pre-electoral coalition formed by the Social Democratic Party (PSD) and the much smaller Conservative Party (PC) topped the polls with 40% of the vote and started preparations to form a new government. Two weeks later, when the candidate endorsed by the centre-right Justice and Truth Alliance (DA) won the presidential runoff, things took a different turn. Although DA had come second in the legislative election, newly elected President Băsescu nominated a PM candidate from his supporting camp as he convinced the PC to break their pre-election agreement with the PSD and join the new centre–right government. In this case, the president successfully affected the formation outcome by bringing the coalition he preferred to a majority status. The Romanian example suggests that certain circumstances may allow presidents to get involved in coalition negotiations and change the government formation outcome that parliamentary parties would have otherwise reached. Arguably, things would have played out differently if the Romanian constitution did not grant the head of state a proactive role in the nomination of the PM. However, the president’s first-move advantage would have made little difference if PSD had won the general election by a greater margin. This is what happened after the 2012 general election, when President Băsescu had little choice but to appoint the PSD leader as PM. This time around, the pre-election coalition between PSD, National Liberals (PNL), and PC won the election with more than 60% of the vote. Under these circumstances, the president was no longer able to use his unconstrained PM nomination power to alter the balance of forces established by the general election.

The evidence regarding non-formateur parties is encouraging as well. In parliamentary systems, while a positive effect on portfolio allocation can be found at both high and low levels of presidential power at small values of ENPP (i.e. below 3.5), the positive effect in conditions of high legislative fragmentation disappears when the president is weak. Similarly, in semi-presidential systems we find a positive effect on junior coalition parties allied with both strong and weak presidents at low and medium levels of legislative fragmentation (i.e. ENPP lower than 4.5). However, for higher levels of ENPP, being a weak president’s party turns into a significant disadvantage.

Conclusion

Shugart and Carey’s (1992) seminal book directed scholars’ attention to the various ways in which fine-grained distinctions between democratic regime types and variation in presidential powers affect executive-legislative relations. While their analysis did not particularly focus on the cabinet, this article has drawn on the implications of their work to study the role of presidents in government formation in parliamentary and semi-presidential democracies. Specifically, the article addressed the ‘portfolio allocation paradox’ (Warwick and Druckman, 2006) in non-presidential systems and proposed that the role of heads of state in government formation be incorporated as an institutional constraint into models of inter-party negotiation over coalition formation. Scholars of parliamentary and semi-presidential systems are increasingly drawing attention to the bargaining power and coalitional potential of the president’s party in explaining coalition outcomes (see, for example, Kang, 2008, 2009; Anghel, 2017; Bucur and Cheibub, 2017; Savage, 2018). To date, however, the question of whether presidents act as partisan actors in government formation and favour their political allies when it comes to the distribution of cabinet spoils has not been thoroughly explored.

The analysis has focused on variation in presidential power to influence cabinet appointments and the context of legislative bargaining over the formation of executive coalitions. The results confirmed that the allocation of cabinet portfolios is roughly proportional in non-presidential systems. Moreover, the allocation of cabinet seats does not seem to differ significantly between parliamentary and semi-presidential systems, despite the separation of powers element introduced by direct presidential elections. However, these results do not necessarily run against the emphasis that Shugart and Carey’s Presidents and Assemblies, and subsequently Samuels and Shugart (2010), placed on the consequences of separate origin and survival for executive-legislative relations. As the semi-presidential literature highlights, as long as no assessment of presidential powers is carried out, the election mode remains a poor indicator of presidential influence on political outcomes broadly defined (Elgie, 2015). In fact, as Samuels and Shugart (2010: 15–16) underline, the impact of the separation of powers on cabinet politics is relevant only ‘to the extent that the capture of a separately elected presidency is important for control over the distribution of the spoils of office and/or the policy process’. Thus, the direct election of the president alone is unlikely to affect executive-legislative relations. Indeed, we find that only in the case of formateurs allied with strong presidents generally, or presidents formally empowered to play a role in government formation, a portfolio bonus seems more likely to emerge. Finally, the results indicate that, while constitutions may carve an important role for heads of state in government formation, the complexity of the bargaining environment also matters for the extent to which presidents can use their formal powers to influence negotiations over cabinet formation.

These results suggest several future avenues for research. First, more work is needed to explore the causal mechanism underlying the presidents’ influence on the distribution of ministerial spoils in non-presidential systems. The increase in bargaining power associated with a strong presidency naturally leads to expectations that, to the extent that deviations from proportionality exist, they advantage presidential allies. That said, we need a more fine-grained understanding of the way in which alliance with a president increases the agenda-setting power of the cabinet and lowers the costs of coopting coalition partners, allowing formateur parties to extract a portfolio bonus. Second, given the common logic of coalition-building in presidential and parliamentary systems (Alemán and Tsebelis, 2011: 25), the explanatory value of presidential powers for cabinet proportionality should be tested in comparisons across democratic regime types. By and large, the literature studying cabinet coalescence in presidential systems and portfolio allocation in parliamentary and semi-presidential regimes expects the formateur advantage to vary with the chief executive’s institutional prerogatives and constraints. In some countries, formateur parties, due to their large size and broad powers, need not be concerned about attracting coalition partners and/or obtaining a working majority. In other countries, formateurs rely on a working majority to implement their programmes, which gives potential partners considerable leverage to extract more portfolios than their proportional share. In between there is a wide spectrum of portfolio allocation mechanisms, which we would expect to be related to political incentives, cabinet formation power, and the complexity of the bargaining environment.

Supplemental Material

PSW856875_Supplemental_material – Supplemental material for Presidents and Cabinet Payoffs in Coalition Governments

Supplemental material, PSW856875_Supplemental_material for Presidents and Cabinet Payoffs in Coalition Governments by Cristina Bucur in Political Studies Review

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Supplemental Material

Supplemental material for this article is available online.

Notes

ORCID iD

Cristina Bucur

References

Alemán

Tsebelis

(2011) Political Parties and Government Coalitions in the Americas. Journal of Politics in Latin America 31: 3–28.

Amorim Neto

(2002) Presidential Cabinets, Electoral Cycles, and Coalition Discipline in Brazil. In: Morgenstern

Benito Nacif (eds) Legislative Politics in Latin America. New York: Cambridge University Press, pp.48–78.

Amorim Neto

(2006) The Presidential Calculus. Executive Policy Making and Cabinet Formation in the Americas. Comparative Political Studies 394: 415–440.

Amorim Neto

Samuels

(2010) Democratic Regimes and Cabinet Politics: a Global Perspective. Revista Iberoamericana De Estudios Legislativos 11: 10–23.

Amorim Neto

Strøm

(2006) Breaking the Parliamentary Chain of Delegation: Presidents and Non-partisan Cabinet Members in European Democracies. British Journal of Political Science 364: 619–643.

Anghel

(2017) ‘Why Can’t We Be Friends?’ The Coalition Potential of Presidents in Semi-presidential Republics – Insights from Romania. East European Politics and Societies and Cultures 2010: 1–18.

Arter

(1999) Finland. In: Elgie

(ed.) Semi-presidentialism in Europe. Oxford: Oxford University Press, pp. 48–66.

Bäck

Dumont

(2008) Making the First Move: A two-stage Analysis of the Role of Formateurs in Parliamentary Government Formation. Public Choice 135: 353–373.

Baron

Ferejohn

(1989) Bargaining in Legislatures. American Political Science Review 834: 1181–1206.

10.

Bassi

(2013) A Model of Endogenous Government Formation. American Journal of Political Science 574: 777–793.

11.

Bischof

(2017) New Graphic Schemes for Stata: Plotplain and Plottig. Stata Journal 173: 748–759.

12.

Browne

Franklin

(1973) Aspects of Coalition Payoffs in European Parliamentary Democracies. American Political Science Review 672: 453–469.

13.

Browne

Frendreis

(1980) Allocating Coalition Payoffs by Conventional Norm: An Assessment of the Evidence from Cabinet Coalition Situations. American Journal of Political Science 244: 753–768.

14.

Bucur

(2017) A Mould-breaking Cabinet? Changes and Continuities in the Formation of the 2017 French Government. French Politics 153: 340–359.

15.

Bucur

(2018) Cabinet Payoffs in Coalition Governments: A Time-Varying Measure of Portfolio Importance. Party Politics 242: 154–167.

16.

Bucur

Cheibub

(2017) Presidential Partisanship in Government Formation: Do Presidents Favor their Parties when they Appoint the Prime Minister? Political Research Quarterly 704: 803–817.

17.

Carroll

Cox

(2012) Presidents and Their Formateurs (Working paper). Available at: http://ssrn.com/abstract=1905305 (accessed 15 May 2019).

18.

Chaisty

Nic

Power

(2014) Rethinking the ‘Presidentialism Debate’: Conceptualizing Coalitional Politics in Cross-Regional Perspective. Democratization 211: 72–94.

19.

Chaisty

Nic

Power

(2018) Coalitional Presidentialism in Comparative Perspective. Oxford: Oxford University Press.

20.

Cutler

De Marchi

Gallop

, et al. (2016) Cabinet Formation and Portfolio Distribution in European Multiparty Systems. British Journal of Political Science 461: 31–43.

21.

Döring

Manow

(2016) Parliaments and Governments Database (Parlgov): Information on Parties, Elections and Cabinets in Modern Democracies. Development Version. Available at: http://www.parlgov.org/ (accessed 15 May 2019).

22.

Elgie

(2015). Heads of State in European Politics. In: Magone

(ed.) Routledge Handbook of European Politics. London: Routledge, pp. 311–327.

23.

Elgie

(2016) Three Waves of Semi-Presidential Studies. Democratization 231: 49–70.

24.

Elgie

Passarelli

(2018) Presidentialisation: One Term, Two Uses – Between Deductive Exercise and Grand Historical Narrative. Political Studies Review 17: 115–123.

25.

Falcó-Gimeno

Indridason

(2013) Uncertainty, Complexity, and Gamson’s Law: Comparing Coalition Formation in Western Europe. West European Politics 361: 221–247.

26.

Glasgow

Golder

(2011) Who ‘Wins’? Determining the Party of the Prime Minister. American Journal of Political Science 554: 937–954.

27.

Kang

(2008) Government Formation and Termination in European Democracies with Presidential Heads of State. PhD Thesis, University of Rochester, New York.

28.

Kang

(2009) The Influence of Presidential Heads of State on Government Formation in European Democracies: Empirical Evidence. European Journal of Political Research 484: 543–572.

29.

Laakso

Taagepera

(1979) ‘Effective’ Number of Parties: A Measure with Application to West Europe. Comparative Political Studies 121: 3–27.

30.

Laver

Schofield

(1990) Multiparty Government. the Politics of Coalition in Europe. Ann Arbor, MI: University of Michigan Press.

31.

Morelli

(1999) Demand Competition and Policy Compromise in Legislative Bargaining. American Political Science Review 934: 809–820.

32.

Müller

(1999) Austria. In: Elgie

(ed.) Semi-presidentialism in Europe. Oxford: Oxford University Press, pp. 22–47.

33.

Protsyk

(2005) Prime Ministers’ Identity in Semi-presidential Regimes: Constitutional Norms and Cabinet Formation Outcomes. European Journal of Political Research 44: 721–748.

34.

Samuels

(2009) Separation of Powers. In: Boix

Stokes

(eds.) The Oxford Handbook of Comparative Politics. Cambridge: Cambridge University Press, pp. 704–26.

35.

Samuels

Shugart

(2010) Presidents, Parties, and Prime Ministers: How the Separation of Powers Affects Party Organization and Behavior. Cambridge: Cambridge University Press.

36.

Savage

(2018) How and When do Presidents Influence the Duration of Coalition Bargaining in Semi-presidential Systems? European Journal of Political Research 572: 308–332.

37.

Schleiter

Morgan-Jones

(2009) Party Government in Europe? Parliamentary and Semi-presidential Democracies Compared. European Journal of Political Research 485: 665–693.

38.

Schleiter

Morgan-Jones

(2010) Who’s in Charge? Presidents, Assemblies, and the Political Control of Semipresidential Cabinets. Comparative Political Studies 4311: 1415–1441.

39.

Seki

Williams

(2014) Updating the Party Government Data set. Electoral Studies 34: 270–279.

40.

Shugart

(2005) Semi-Presidential Systems: Dual Executive and Mixed Authority Patterns. French Politics 33: 323–351.

41.

Shugart

Carey

(1992) Presidents and Assemblies: Constitutional Design and Electoral Dynamics. Cambridge: Cambridge University Press.

42.

Strøm

Budge

Laver

(1994) Constraints on Cabinet Formation in Parliamentary Democracies. American Journal of Political Science 382: 303–335.

43.

Tavits

(2008) Presidents with Prime Ministers: Do Direct Elections Matter?. Oxford: Oxford University Press.

44.

Urbanavicius

(1999) Lithuania. In: Elgie

(ed.) Semi-presidentialism in Europe. Oxford: Oxford University Press, pp. 150–169.

45.

Warwick

Druckman

(2001) Portfolio Salience and the Proportionality of Payoffs in Coalition Governments. British Journal of Political Science 314: 627–649.

46.

Warwick

Druckman

(2006) The Portfolio Allocation Paradox: An Investigation into the Nature of a Very Strong but Puzzling Relationship. European Journal of Political Research 454: 635–665.

47.

Williams

Seki

(2016) Seki-Williams Government and Ministers Data. Harvard Dataverse, Version 3.0. Available at: https://doi.org/10.7910/DVN/0UNUAM (accessed 15 May 2019).

48.

Woldendorp

Keman

Budge

(2000) Party Government in 48 Democracies (1945–1998): Composition, Duration, Personnel. Dordrecht: Kluver Academic Publishers.

49.

Zarate’s Political Collections (2015) World Political Leaders, 1945–2015. Available at: http://zarate.eu/countries.htm (accessed 15 May 2019).

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.26 MB