[ \ \ \\
|   Institutional Design and Party Government in \ Post-Communist Europe
Csaba Nikolenyi
1
The Institutional Sources of Minority Governments
5.1 INTRODUCTION
Held under an extremely permissive and proportional electoral system in October 1991, the first post-communist parliamentary elections in Poland produced what became to this day the most fractured and fragmented legislature in all of East Central Europe. With the largest party controlling no more than 13.5 percent of the seats in the Sejm, it was not very surprising that political parties found it extremely difficult to put together a majority coalition that would be capable of looking after crucial legislative business involving social, economic and political reforms. Indeed, after two successive minority coalition governments proved unable to maintain majority support on the floor, the Sejm was dissolved, prematurely paving the way for fresh elections in 1993.
The same year Bulgaria also held the first post-transition elections to the Narodno Sabranie. Relative to Poland, the first Bulgarian parliament was considerably less fragmented in its composition than the Polish Sejm: only three political parties won legislative seats with the largest among them accounting for over 40 percent of the newly elected deputies. Similarly to the Polish case, however, the largest party, the Union of Democratic Forces (SDS), also opted to form a minority government, which, in retrospect, ended up being only slightly more durable than the minority coalitions in the Polish Sejm.
These examples suggest that political parties in the early post-transition legislatures of post-communist East Central Europe did not easily succeed, or were willing, to comply with the dicta of conventional coalition theoretic wisdom to form majority-size governments that either maximize their participants' office-related payoffs (Riker 1962), minimize the ideological distance among the coalition's partners, or both (Axelrod 1971). Instead, parties often opted to form undersized minority governments. Echoing earlier, and largely West European, explanations of minority governments (Taylor and Herman 1971), these examples suggest that undersized governments may well be the product of unpredictable unique, or crisis, situations in which the normal logic and dynamics of coalition bargaining are more difficult to apply.
The Institutional Sources of Minority Governments 89
Indeed, we might plausibly claim that the first post-transition elections of the respective new democracies presented parties with precisely the kind of unpredictability of circumstances to which these conventional "crisis" explanations attribute the formation of minority governments. From this perspective, it may not be very surprising to find that these first post-communist governments tended to be under-sized. Moreover, the well-known under-institutionalization of the party systems in new democracies might also seriously weaken the ability of political parties to carry out and implement the types of calculations that would lead to the formation of both majority-size coalition governments. Thus, in the case of the early post-transition East Central European legislatures, the formation of undersized governments may be merely reflective of the still nascent party systems where high electoral volatility and the high incidence of new party formation makes the precise calculation of the various coalition combinations particularly difficult.
If the formation of these early minority governments were truly the result and the artifact of historical time and context, then over the next several elections we should have witnessed post-communist parties conforming increasingly more to the logic and rationality of office- and policy-seeking and, as a consequence, fonning majority governments as their national party systems stabilize and mature. If so, then the rate and frequency of minority governments in post-communist East Central Europe should also be very different from what we find in the established democracies of Western Europe, where "minority solutions have been chosen 42.8 percent of the time" and "[t]he frequency of minority governments has increased overtime" (Strom 1990: 59). Indeed, we can see that whereas half of the ten new democracies witnessed the formation of a minority government after their first post-transition election (Bulgaria, Czech Republic, Latvia, Poland and Romania), the subsequent rate of minority cabinet formation after parliamentary elections declined to 18 percent (only eight of the forty-four subsequent elections up to 2010 resulted in minority governments). Yet, considering both post-election and mid-term governments, the frequency of minority cabinets is high: in approximately half of the instances that a new government is formed, either after a parliamentary election or during the same inter-election period, political parties have formed under-sized minority governments.
Whether the first post-transition government in a new democracy was undersized or not seemed to have an effect on the frequency of minority governments in the future: all but one of the five states where the first post-communist election led to a majority government (Estonia, Hungary, Lithuania, Slovakia, and Slovenia) continued the same pattern after each subsequent parliamentary election. The only exception is Lithuania but even there we find only a single instance of a postelection minority cabinet in the year 2000. In contrast, six of the seven minority governments that were formed after later elections came about in those new democracies where the first post-transition election already produced such an
The Institutional Sources of Minority Governments
1
The Institutional Sources of Minority Governments
91
outcome: two more elections led to minority cabinets in the Czech Republic and Romania respectively, and one more in Bulgaria, Latvia, and in Poland.
Important and relevant as context-specific explanations might be, the cross-sectional distribution of minority governments suggests that there are additional factors at work that may account for variation in the formation of under-sized governments across the new democracies. Why is it the case that some post-communist democracies never witness the formation of a minority government after a general election (e.g. Estonia or Hungary) while in other cases such an outcome is almost a norm rather than the exception (e.g. Czech Republic or Romania)? Indeed, minority governments in general arc concentrated in four states: the Czech Republic, Poland, Romania, and Latvia; it was not until very recently that political parties started to arrive at minority cabinet solutions again in Bulgaria. An obvious candidate answer to this question would be the incentive structure provided by the prevailing institutional design of the new democracies that define both the rules of the government formation game and the constraints under which it would be played out. With respect to the institutional context of government formation, the ten new democracies have indeed shown sufficiently h igh variation which makes further analysis of their impact both worthwhile and justified. Specifically, the chapter will argue that institutional features that hinder the concentration ofpower in, and as such weaken the relative authority of the first chamber ofparliament over the executive, also reduce the incentives for political parties to form majority coalitions. As discussed earlier, institutional features that promote the dispersion of political power include a large assembly size, bicameralism, negative parliamentary government formation rules, and a strong presidency.
This institutonalist argument about minority government formation stems from the central assumption that minority governments are formed for two reasons. On the one hand, there are potential members of a governing coalition whose entry could increase the size of the coalition to a majority but that decide to stay outside the government. On the other hand, it is also equally likely and possible that those members that are already inside the coalition choose not to increase the size of the coalition any further because they calculate that they would be able to govern from a minority position. Both of these calculations are shaped by the incentives provided by the prevailing institutional structures, specifically the limitations on the authority of the first parliamentary chamber. When the first chamber is the single source of authority of policy making and when there are few limitations on its overall political power, parties have a strong incentive to form majority coalitions that capture and control it. Conversely, when the authority and the powers of the chamber are more limited, the incentive to form a winning coalition becomes weaker as both the opposition and the governing parties may calculate that they can exercise control over policy and legislation through their access to those additional offices and sites of power that limit the authority of the first chamber. The institutions that may limit the overall power of the first chamber are its relatively large size, bicameralism, presidential power, and a negative formation rule.
The chapter starts with an overview of the main arguments in the literature about the causes for the formation of minority government with particular emphasis on the role of institutional arrangements. This will be followed by a comparative summary assessment of the history of government coalitions in the ten states. A detailed country-by-country review is provided in Appendix B. The third section evaluates the role of political institutions.
5.2 EXPLANATIONS OF MINORITY GOVERNMENTS
According to a long tradition of research on party systems and patterns of government formation, minority governments are associated with high degrees of fragmentation, polarization and instability in the party system, which reduce both parties' willingness to bargain with each other to form a majority coalition and the certainty of the information that they have about each other's strategies, preferences and goals (Uodd 1976). In this vein, Sartori argues that minority governments are particularly characteristic features of "moderate pluralism," a party system format defined by moderate degrees of fragmentation, with three to five relevant parties, and ideological polarization. In such systems, "minority single party governments do materialize, but they do so either as a result of miscalculated Indian wrestlings, or on the basis of a precise calculus (such as shedding unpopular, if necessary, policies), and otherwise as disguised coalitions and transitional caretaker governments" (Sartori 1976: 178).
A different line of explanation predicts that minority governments will be formed by a party that is located in the center of the issue space of the party system. Building on Black's (1958) well-known median voter theorem, which holds that the winset of the median player is empty in one dimension under majority rale, a number of authors have argued that a centrally located party can divide the majority opposition and form a minority government on its own (van Roozendaal 1992a,b; Crombez 1996; Laver and Shepsle 1996; Schofield 2007). By forming a government on its own rather than sharing office with its coalition partners, the center party can maintain control of the portfolio allocation process, which allows it to maximize its office benefits. At the same time, since the winset of the median player is empty, by forming a minority government the center party can also maximize the likelihood that government policy would reflect its own ideal point.
In this vein, Laver and Shepsle (1996) propose that a minority government will be formed by a very strong party, which is characterized by having "an ideal point such that there is no alternative government preferred by a majority to one that gives the very strong party all portfolios" (1996: 263). Van Roozendaal (1992a,b) predicts that the likelihood of both the formation and the stability of a minority
92 The Institutional Sources of Minority Governments
government increase when the central party is also numerically dominant. Crombez (1996) arrives at essentially the same conclusion and predicts that the larger and more centrist the plurality party the greater the likelihood that it would form a minority government.
A more dynamic explanation of minority governments is offered by Grofman, Straffin and Noviello (1996) who model cabinet formation as a sequential process of proto-coalition formation. According to this model, a minority government will have a knife-edge quality to it in that it will be formed only when two proto-coalitions have reached equal size and no further expansion is possible. In a sense, this model provides a theoretical foundation for earlier obseivations by Taylor and Laver (1973) and Herman and Pope (1973) according to which most minority governments, in Western Europe, are formed by near-majority size parties because a "party, like a government, which controls over forty-five percent of the seats in a legislature is likely to receive the support necessary to provide it with a majority on crucial decisions.... This suggests that the effective decision point in these parliaments is a number smaller than a simpler majority" (Herman and Pope 1973: 203)
A third approach to understanding the formation of minority governments stresses the institutional incentives that encourage office- and policy-seeking political parties not to enter executive office and allow the formation of a minority cabinet instead of a majority coalition (Strom 1990; Laver and Budge 1992; Bergman 1993; Strom, Budge and Laver 1994). A number of different explanations have been proposed in this new-institutionalist vein. For example, Bergman (1993) links the formation of minority governments to negative parliamentary rules of cabinet formation, while Strom, Budge and Laver (1994) point out those restrictive legislative rules favor minority coalitions, while provisions for mandatory government size, as in the case of constructive no-confidence votes, rule out the formation of a minority government.
The most complete and influential account of minority government formation in the neo-institutional perspective is provided by Shorn (1984, 1990) who identifies two institutions that provide incentives for political parties not to enter office: the influence of the parliamentary opposition on policy and the electoral decisiveness of government formation. The stronger the role of the opposition in the policy making process, the greater the likelihood that policy-seeking potential coalition partners may want to stay outside the formal structure of the executive in order to avoid incurring the electoral costs of incumbency. This disincentive to enter executive office is further exacerbated where electoral outcomes are decisive of coalitional bargaining power: instead of entering office now, which is costly in electoral terms, parties may strategically calculate that it is better to wait until their electoral chances will allow them to enter office on more advantageous terms in the future and let someone else incur the costs of incumbency in the meantime.
The literature on party coalitions in post-communist democracies has paid scant attention to the issue of minority governments. Instead, scholars have focused on
The Institutional Sources of Minority Governments 93
how the regime divide, i.e. the bipolar competition between successors of the former communist parties and their opposition, has constrained parties' coalitional choices after the transition by imposing an electoral penalty on parties that cross this divide (Grzymala-Brusse 2001). An important finding in this stream of the literature is that coalitions that include communist successor parties tend to be oversized and that such parties tend to receive a less-than-proportional share of cabinet portfolios compared to other parties (Druckman and Roberts 2007). Other studies have pointed to the impact of institutional factors, such as the role of the president in appointing the formateur (Protsyk 2005), and the presence of dominant and central parties in the legislature (Nikolenyi 2004b) in affecting cabinet composition and duration.
One reason for the lack of significant scholarly interest in the study of post-communist minority governments may be that "[sjince the new parties in the region have been initially described as having vague ideologies, few clear policy differences, and office as their main goal, there are ... grounds to expect that the coalition in the region will follow the minimum winning coalition model" (Grzymala-Busse 2001: 86). If so, then minority governments are mere aberrations rather than patterned outcomes systematically and consistently generated by particular features of the new democratic political systems. Indeed, as we shall see in the following section, the majority of post-communist governments were of a majority size, and minority governments mainly proliferated in the first post-transition parliaments when policy constraints on coalition bargains were much weaker than later on as the party systems were becoming more institutionalized and ideologically patterned. Thus, although the lack of interest in post-communist minority governments is not entirely unjustified, it docs create a lacuna in our current understanding of coalition dynamics in these new democracies.
5.3 MINORITY GOVERNMENTS IN POST-COMMUNIST DEMOCRACIES
Before providing a detailed description about minority governments in the new democracies a proper definition of the dependent variable is in order. In his seminal work on minority governments in parliamentary democracies, Strom defines a majority cabinet (or government) as "any cabinet that meets all appropriate constitutional requirements and that is composed of persons acting as representatives of political parties or parliamentary groups that collectively control no less than half of all seats in the national legislature, or that chamber of the legislature to which the cabinet is constitutionally responsible" (Strom 1990: 6). In turn, he defines minority cabinets (or governments) as those that "meet all of the foregoing requirements except the majority clause." Strictly speaking, this
<)4
The Institutional Sources of Minority Governments
The Institutional Sources of Minority Governments
93
definition includes cabinet coalitions at the knife-edge, (Grofman, Strafrin and Noveillo 1996), that is to say at exactly the 50 percent mark, among majority governments. Strom justifies this operational decision by explaining that "most parliaments have rules favoring the status quo in the event of tie votes. In parliamentary regimes this generally means the party or parties in office. Even in the absence of such a bias, the worst predictable outcome for the government would be a stalemate. But under no existing institutional rule would this suffice to bring down the incumbent government. And the crucial property of a minority government is precisely that the composition of parliament appears sufficient to bring down the government at any time". The formation of government coalitions rarely stops at the knifc-edge, however, it does happen from time to time. In our dataset of post-communist cases, wc find one instance of such a government, the Topolanek cabinet formed after the 2006 Czech parliamentary elections. Even in the established parliamentary democracies of Western Europe we can only find a handful of such instances in post-war Western Europe (in Italy, Ireland, Iceland, and Sweden).
A second operational question that needs to be discussed pertains to the exact nature of coalition partners' commitment to the government. Sometimes, parties may consider it prudent to commit only their external support to a minority government without formally accepting ministerial portfolios. The case of the Opposition Pact between the Czech Social Democratic Party and the ODS is perhaps the best known example of this in our sample (see Nikolenyi 2003); following the 1998 parliamentary election, the two parties entered into a formal contract which allowed the formation of a single-party CSSD minority government in exchange for ceding key legislative offices, such as the Speakership of the lower house and the chairmanship of important committees, to the largest opposition party, the ODS. Other instances where a party was formally committed to a governing coalition but to not to the government per se, which thus remained in a formal minority, include the Kristopans government of Latvia of 1998 (Davies and Ozolins 2001) and the Nastasc government of Romania in 2000 (Popescu 2003). I agree with Strom that a party's commitment to a government is credible only if it is followed up by delegating at least one minister to the government; therefore, agreements of external support, including instances of coalitional membership that are not backed up by portfolio allocation, are not considered in our view as instances of partnership and membership in government. Therefore, I count each of the three instances above as cases of minority governments even though they were based on a negotiated majority coalition in the legislature.
The third issue relates to the presence of a majority party in the legislature, which effectively precludes the possibility that a minority government might form. In the language of cooperative game theory, majority parliaments constitute nonessential games, where the election has already produced a winner. Therefore, although we may witness the formation of oversized surplus majority coalitions
Table 5.	Minority governments in post-communist East Central Europe, 1990 -2010		
State	Number of	Number of minority	Number of post-election minority
	elections	parliaments (%)	governments (% of minority parliaments)
1, Most favorable institutional conditions tor power concentratioi			
Estonia	5	5 (100)	0
Hungary	6	4 (66.7)	0
Bulgaria	6	4 (66.7)	2
Slovenia	5	5 (100)	0
Group average	5.5	4,5 (81.8)	9.5 (11.1)
2. Mixed institutional conditions for power concentration			
Latvia	6	6(100)	2
Slovakia	5	5 (100)	0
Group average	5.5	5.5 (100)	1 (18.2)
3. Least favorable conditions for power concentration			
Lithuania	5	4(80)	1
Poland	6	6 (100)	2
Romania	5	5 (100)	3
Czech Republic	5	5 (100)	3
Group average	5.25	5 (95.2)	2.25 (45)
Total	54	49 (90.7)	13 (24.1)
Note: The numbers in brackets indicate average values in %.
even if a single patty has won a legislative majority, such coalitions do not serve the immediate purpose of winning the government formation game per se. Of the fifty-four elections that were held in the ten post-communist democracies between 1990 and 2010 only five resulted in a majority winner: Hungaty 1994 and 2010, Lithuania 1992, Bulgaria 1994 and 1997. In all but one instance, the exception being Hungary 1994, the majority party formed a government by itself.
Using these operational definitions, Tabic 5.1 shows the distribution of minority governments across the ten new democracies.
As mentioned above, the overwhelming majority (90.7 percent) of post-communist elections did not produce majority winners. This is hardly surprising given that none of these new democracies adopted plurality or majority electoral systems that are normally associated with the election of majority parliaments (Rae 1967; Lijphart 1994). As we saw in Chapter Two, with the exception of Hungary and Lithuania, which adopted mixed-member electoral systems, all other post-communist democracies use different forms of proportional representation. Since mixed-member systems incorporate a plurality/majority tier, it is hardly surprising that we actually find more than half of the majority election outcomes (three out of five) in these two states; also two of the remaining instances of majority election results are round in Bulgaria, which, interestingly, is also the state with the lowest magnitude PR system. It is well known from tire literature on the political
96
The Institutional Sources of Minority Governments
T
consequences of electoral laws that district magnitude has a positive effect on the degree of parliamentary fractionalization, which, however, has a negative effect on the likelihood of the formation of majority governments, ceteris paribus (Rae 1967; Lijphart 1994). Therefore, the election of so many majority parliaments in Bulgaria may not be so surprising.
Almost exactly one-fourth (thirteen out of forty-nine) of the elections that produced a minority parliament were followed by the formation of a minority government. While most majority governments are formed by multiparty coalitions, almost half of the minority governments (five out of thirteen) are formed by single parties, in each case by the party that won the most seats in parliament after the election. The five single-patty minority governments are as follows: Bulgaria 1991, Czech Republic 1998, Poland 2005, and Romania 1992 and 2000.
In Western Europe, we find that minority governments are geographically concentrated in Scandinavia. Strom reports that sixty-one out of the 125 postwar minority cabinets were formed in the five Nordic democracies: Denmark, Finland, Iceland, Norway, and Sweden (1990: 58). In the new East Central European democracies, we do not find such a clear regional pattern. The three Baltic states (Estonia, Latvia, and Lithuania.) account for three, while the five Central European states (Czech Republic, Hungary, Poland, Slovakia, and Slovenia) and the two Balkan democracies (Bulgaria and Romania) have had five post-election minority governments respectively. As individual countries, the Czech Republic and Romania stand out from the group by virtue of having no fewer of their elections followed by the formation of a minority cabinet than by a majority government.
Table 5.1 clearly suggests that the institutional concentration/dispersion of political power is related to the frequency of minority governments to form after the general elections: as we move from the first down to the third group of states the rate of post-election minority governments, i.e. the number of post-election minority governments divided by the number of all minority parliaments in the group, increases sharply from 11.1 percent in the first group (two out of eighteen), to 18.2 percent in the intermediate (two out of eleven), and reaches 45 percent in the third group (nine out of twenty). In other words, as the institutional dispersion of power increases so does the incidence of minority cabinets. Within the first and the last groups, however, there are anomalies to draw attention to. In the first group Bulgaria alone accounts for all post-election minority governments. Moreover, the number of these cabinets in Bulgaria is exactly the same as in those in Latvia, which belong to the next group. In the last group, Lithuania and Poland seem to deviate as they have similar numbers of post-election minority governments as states in the first two groups with high and intermediate degrees of power dispersion.
The Institutional Sources of Minority Governments 97
5.4 AN 1NSTITUTIONALIST ACCOUNT OF POST-COMMUNIST MINORITY GOVERNMENTS
I propose six specific hypotheses in order to evaluate the effects of institutional design on minority government formation.
Hypothesis la: District magnitude is positively related to the likelihood of minority government formation. As discussed above, party system fragmentation has been one of the reasons that conventional wisdom held responsible for minority governments. Since average district magnitude is well known to shape the number of parties in a negative way, i.e. the smaller the district magnitude the fewer the parties, it is reasonable to expect that district magnitude would have an indirect positive impact on the probability that a minority government is formed after an election. I will also assess the direct impact of the number of parties, measured by the Laakso-Taagepera index of the effective number of parliamentary parties, on minority governments.
Hypothesis 2: Large assemblies promote the formation of minority governments. As discussed earlier, assembly size inversely affects the concentration of political power. Since power gets more diffused as the number of scats increase, parties may be less likely to be motivated to invest in entering the executive. In addition, as the number of seats increases the sheer operational cost of coordinating more politicians around government formation increases. Simply put, it is easier to get fewer deputies together and form a majority coalition in smaller parliaments and ensure executive control over the legislature (Lijphart 1999). Hypothesis 3: Bicameralism promotes the formation of minority governments. In contrast with other scholars (Lijphart 1984; Diermeier, Eraslan and Merlo 2007), I expect that bicameralism promotes the formation of minority governments either by virtue of limiting the policy authority of the first chamber, which reduces the benefit of government participation for prospective and potential coalition partners, or by encouraging the party that leads the government to keep power-sharing to the minimum if it has acquired a numerically strong position in the second chamber. (For the purposes of this analysis, I consider Slovenia de facto unicameral because the unelected nature of the National Council does not reduce the policy authority of the first chamber in any appreciable way.)
Hypothesis 4: Constitutional provisions that provide a positive role for the legislature in the governmentformation process promote the formation of majority coalition governments while negative formation rules promote the formation of minority governments (Bergman 1993). Under positive parliamentary formation rule, the party that is appointed to lead the government formation process needs to build a majority coalition before it can be invested in office. Under negative parliamentary rules, a government can be invested so long as there is no hostile majority coalition that can prevent it from passing a vote of confidence.
98 The Institutional Sources of Minority Governments
Hypothesis 5: Strong presidencies promote the likelihood of minority government formation. It is well known that minority governments abound in presidential systems of government (Cheibub 2007). Although none of the post-communist democracies have a presidential constitution, strong chief executives can promote the formation of minority cabinets for similar reasons as they do in presidential systems of government: since control over executive authority becomes increasingly more shared and contested between the legislature and tire presidency as the latter becomes more powerful, political parties will have weaker incentives to focus their power-seeking efforts on establishing political control over the legislature.
Hypothesis 6: The. likelihood of minority government decreases after the first-post communist election. As mentioned at the outset of the chapter, an important characteristic of post-communist minority governments is their temporal concentration in the first post-transition legislatures. It is not unimaginable that due to the novelty of representative and competitive democracy, parties may take longer to learn the art of building majority solutions in a recurring fashion. This view is supported by the descriptive fact that while half (five out of ten) of the first post-transition elections in the respective states (Bulgaria 1991, Czech Republic 1996, Latvia 1993, Poland 1991, and Romania 1992) produced a minority government, only eight out of the forty-four subsequent elections did so.
I evaluate these hypotheses first by running a number of bivariate logistical regressions; the results are reported in Table 5.2. Most of the variables are defined dichotomously as dummies. For the variable "First parliaments" I code 1 for every observation of a government that is formed after the country's first post-communist election and 0 for every other later election. The next three variables, i.e. Large assembly, Bicameralism, and Positive investiture are defined exactly the same way as they were in the previous chapter. The variable Strong president is the inverse of the Weak president variable from Chapter Three: for each election held in a state where the presidency is powerful 1 assign a value of 1 otherwise a value of 0. Assembly size is defined as the actual number of seats in the first parliamentary chamber at the time of the election. Number of parties is defined as by the effective number of parliamentary parties index proposed by Laakso and Taagepera (1979).
The bivariate equations show that the impact of most of the institutional variables on post-election minority government formation is in the expected direction. Thus, holding all else equal, minority governments are more likely to be formed after the first post-communist election; in states with a large a parliamentary assembly; in states with a bicameral legislature; and when the constitutional design established a relative more powerful presidency. At the same time, minority governments are less likely when government formation rules call for a positive parliamentary investiture. The only variables that have a statistically significant effect are Large assembly,
The Institutional Sources of Minority Governments 99
Table 5.2 Bivariate logistical regressions of the likelihood of minority governments
Independent variable
First legislature	1.6094* (0.7781)
Small district	0.1005 (0.6674)
District magnitude	-0.0126(0.0124)
Number of parties	0.0748 (0.2079)
Large assembly	1.6560* (0.7418)
Assembly size	0.0029 (0.0024)
Bicameralism	1.7228* (0.6969)
Positive investiture	-0.2978 (0.6530)
Strong president	0.9445 (0.6765)
N	49
Note: " p<0.05. The numbers in brackets indicate standard errors.
Bicameralism and First legislature; these three variables also have the highest odd ratios of minority government formation.
The important exception is district magnitude. When measured by the actual average district magnitude, this variable turns out to have an unexpected negative effect on minority government formation. When measured by a dummy, with districts of fewer than ten average seats coded as 1 and all else coded as 0, the effect of the variable remains in the unexpected direction and retains a modest magnitude. The Number of parties variable has the expected positive but very modest effect on the probability that a minority cabinet is formed after the election.
Table 5.3 reports the results of multivariate logistical regressions, which lead to a number of important observations. First, the Bicameralism variable behaves inconsistently: it has a negative effect on minority government formation in Models Two and Three but a positive one in the other three Models; however, the effect is very small is Model One. It is important to note that all three models where Bicameralism behaves in this unexpected fashion also includes the Large assembly variable. This is a very important point because there is a high overlap of states that have both of these institutions present, i.e. three of the five states with large first chambers (Czech Republic, Poland, and Romania) are also bicameral. Therefore, to ensure that they do not cancel each other out, I do not include them together in Model 4. Indeed, the effect of Bicameralism becomes significant and positive in that Model.
A second important observation is that four variables (First legislature, Large assembly, Positive investiture, Strong president) have a consistent predictable effect on the probability of minority government formation: large assemblies with at least 200 seats in the first chamber make it more likely that the postelection government will be of a minority size, as do relatively more powerful presidencies and the first post-transition elections. Positive government formation rules, on the other hand, have a consistent negative impact on minority
3
60
'C
1
2
I
W t?
CN — O CN -— *o m rn
rf m gs o ■
CO ^ a M
cc pa —
CO © f"j
o r-i —^ Q
Cj> o\ a\ o\
O (N o F*
rN m ro oo
fS V, t ^
© ci © r-i
I I
r-- so —
OS   PS   CN O
1 £• 5
■a '£
g 3 £ S3
3
~ a.
E .3 3 -8
S a, CI ^ S ,
T
The Institutional Sources of Minority Governments
SOI
governments. In sum, institutions clearly have a very strong effect on the likelihood that the first government that parties form after a general election will be of a majority or a minority size. More specifically, since bicameralism, large assemblies, a stronger presidency, and negative government formation rules are institutions that arc designed to disperse political power, it can be concluded that institutional design that reduces the concentration of political power in a parliamentary system will also promote the formation of minority governments. What happens to these cabinets thereafter, how durable they are likely to be is a matter that will be examined in the next chapter.
5.5 CONCLUSION
This chapter has made two important contributions to the study of how institutional design shapes party government in ECE. First, it has shown that political institutions notwithstanding, political patties were more likely to form minority governments early on after the transition to democracy. Overtime, in most states, majority coalition combinations, of both the minimum winning and the oversize majority type, have become the norm. In a small number of states, however, this has not been the case. The second important contribution of this chapter has been to show that the institutional design of power dispersion matters as far as the formation of post-election minority governments arc concerned. Of the institutions reviewed, bicameralism has a particularly interesting effect. It is important to reiterate that although bicameralism comes in different varieties (Lijphart 1999), there are good reasons why political parties may not want to form majority coalitions, where they otherwise could, when the government formation game takes place in a bicameral setting: the second chamber, depending on its partisan constitution, may provide either those parties that arc forming the government and/or those that are in opposition with resources that will reduce their incentive either to expand the coalition or join it in the first place. Either way, the overall legislative authority and political power of the first chamber is weaker in bicameral legislatures relative to unicameral ones, and, therefore, it is not surprising to see that political parties will be much more likely to form minority cabinets there.
What is very surprising, however, is that bicameralism seems to work very differently in Western Europe where such legislatures tend to lead to larger governing coalitions rather than smaller ones. An examination of why this is the case does not fall within the scope of this book. It is quite plausible, however, that bicameralism may have become an instrument of power sharing and consensus (Lijphart 1999) in the established democracies precisely because their party systems are established and stable. In the new democracies, however, parties are less stable, their electoral followings are more volatile, and they may be less
102 The Institutional Sources of Minority Governments
predisposed to power sharing in general, which in turn may be reflected in the preponderance of minority governments under bicameralism. In any event, it remains an important finding that bicameralism in ECE does not seem to be the same instrument of power-sharing as it is in Western Europe but that it actually encourages a more narrowly based coalition of parties to capture and hold onto executive office.
6
The Institutional Sources of Cabinet Duration
6.1 INTRODUCTION
The ten post-communist democracies vary considerably in terms of the stability and duration of their governments. In some states, such as Hungary, the government that is formed and invested in office after the election tends to last for the entire or nearly the entire term of the legislature. Elsewhere, such as Latvia, postelection governments never last for a long time. In this chapter, 1 examine how the set of political institutions that disperse or concentrate political power in the electoral, legislative and executive arenas of party competition affect government duration. The central finding of this chapter is the several institutions that disperse political power also reduce cabinet duration: specifically, bicameralism, a more powerful presidency, and negative government formation rules. In addition, government stability inversely varies with the fragmentation of the legislative party system, and both majority and post-election governments are more stable than minority and later mid-term cabinets.
The organization of this chapter follows the same order as the previous ones. The first section reviews the key arguments in the comparative politics literature about the effects of political institutions on government stability. The second section reviews the data on cabinet stability by using a number of alternative measures of the dependent variable. The third section evaluates a number of hypotheses on government stability.
6.2 EXPLANATIONS OF GOVERNMENT STABILITY AND INSTABILITY
Arguments about the institutional causes of government stability have proceeded at throe hierarchically ordered levels of analysis moving from the most immediate cabinet-level characteristics to more remote regime-specific attributes. At the first level we find arguments that attribute government stability to particular characteristics of the cabinet in power, specifically its size, composition, coalitional
1
104 The Institutional Sources of Cabinet Duration
status, as well as the time of its formation in the context of lifespan of the legislature. In a classic statement that originated the myth about the inefficiencies of multipartism, Lowell argued that only the single-party majority governments that are formed in a two-party system can remain stable, while the coalition and minority governments that characterize multiparty systems are doomed to instability, because by their veiy nature, minority cabinets and coalition cabinets are transient (Lowell 1896). Several decades later Blondel (1968) echoed the same conclusion adding the nuance that single-party majority governments will be stable even if they are formed in two-and-a half or multi-party systems. At the same time, Blondel notes, that "[cjoalition, whether small or large, appears directly antagonistic to stable government..." (Blondel 1968: 199).
Dodd (1976, 1974) and Laver (1974) point out that the distinction between single-party and coalition cabinets provides a less powerful explanation of cabinet stability than the distinction among the types of coalition cabinets that are formed. Dodd specifically proposes that cabinet durability is the function of the minimal winning status of the governing coalition: whereas minimal winning coalition cabinets are stable, minority coalitions and surplus majority coalitions are not. Dodd attributes the inherent stability of minimum winning coalitions to the combination of two characteristics of the bargaining environment that produces governments of this of this type: strong willingness by political parties to bargain and the availability of certain and reliable information about parties' relative policy positions. Thus, oversized coalition cabinets are unstable because their formation reflects a party system in which parties' willingness to bargain is coupled with low information certainty. Finally, undersized cabinets arc unstable because the party system in which they arc formed is characterized by low information certainty and low willingness to bargain (Dodd 1974, 1976). A different explanation links the stability of minimum winning coaliton cabinets to the mutually credible exit threats that every member of such a coalition possesses: if the coalition is minimally winning then every member thereof can threaten to leave and cause the termination of the coaliton (Grofman and Roozendaal 1997: 431).
The empirical record seems to support these arguments. For example, using five different measures of government stability, Lijphart (1984) finds support for both sets of explanations in the context of West European governments: (i) single-party governments are more stable than coaliton governments and that (ii) minimum winning coalitions significantly outlast both minority and oversized cabinets (Lijphart 1984: 275, 276). The relatively shorter duration of minority governments is also confirmed by Strom (1985) although he notes that this docs not mean that minority governments arc transient and irrational eabinet solutions. The studies in Budge and Keman (1990) show that single-party governments are the most stable type of all in Western Europe. However, these studies also show that in some countries minimum winning coalitions may not be more stable than cither minority or surplus majority governments. The significance of distinguishing between within-country
The Institutional Sources of Cabinet Duration 105
and cross-country effects of cabinet stability is stressed most forcefully by Grofman (1989) who claims that the relationship between minimal winning status and cabinet durability is the "artifact of the high average duration of cabinets in countries where there are only two or three significant political parties (where minimal winning coalitions are the norm) and the low average duration of cabinets in countries with a very large number of parties (where minimal winning coalitions are rare)" (Grofman and Roozendaal 1997: 431). Grofman further proposes that it is not the type of cabinet that determines eabinet durability but rather the format of the party system, namely, its fragmentation and polarization, that determine both. Therefore, the relationship between cabinet type and cabinet durability, according to Grofman, is at best spurious.
In addition to the size and coalitional type of the government, three other cabinet-level institutional variables have been linked to cabinet stability. The first is the fragmentation of the cabinet, which originally Taylor and Herman (1971), and Sanders and Herman (1977) showed to have a negative effect on cabinet durability. Subsequent studies have provided strong empirical support for these initial findings. The second cabinet level explanation of stability has to do with the time and the number of attempts it takes to form a government. While the relevance of this variable is generally recognized, the nature of its relationship with cabinet stability is not clearly understood either theoretically or empirically. For instance, Strom (1985) reports that a long drawn out foimation process leads to a stable government because the coalition partners can afford to work out the terms of their cooperation in more detail. On the other hand, Laver and Schofield (1990) claim that a long formation period is indicative of a tense and polarized bargaining environment that, in general is less conducive for cabinet stability. Finally, a third cabinet-specific variable related to the stabilty of the government is whether the cabinet is the first post-election government or not. Grofman and Roozendaal (1997) note that post-election governments tend to be more stable than those that are formed later in the term and that caretaker governments are particularly short-lived.
Moving away from the level of the individual cabinets, the next institutional source of cabinet stability is the format of the parliamentary party system, i.e. the balance of powers among political parties in the legislature. It has been well established in the literature that there is a negative relationship between the fragmentation of the party system and cabinet stability (Lijphart 1984; King etal 1990; Grofman and Roozendaal 1997). However, as mentioned above, Grofman notes that the effects of the party system on cabinet stability may well be indirect and mediated through the format and the type of the cabinet formed. Moreover, Laver and Schofield (1990: 149) point out that while it is true that "countries with bigger party systems have less stable cabinets . . . variation in cabinet stability within individual countries reveals little or no relationship between party system size and cabinet stability."
106
The Institutional Sources of Cabinet Duration
Institutions at the third level are the least likely to change over time and will therefore, have the potential to have the most lasting effect, while both the legislative party system and the characteristics of the cabinet in office may vary in the lighl of the result of every new parliamentary election. At this level we find institutions that regulate the length of the term of the legislatures; the ride of the government formation and termination processes; as well as the nature of legislative organization (Strom, Budge and Laver 1994). In countries where the incumbent government can dissolve the legislature and call fresh elections fairly easily cabinets will be less durable than in those states where early elections are difficult to call. Furthermore, constitutional provisions for a constructive vote of no-confidence in the government may make it more difficult for the opposition to defeat the incumbent on the floor of the legislature, which should, ceteris paribus the level of cabinet stability. With regard to the rules of cabinet formation, Bergman (1993) finds that positive parliamentary rules of government formation, which are defined by a formal investiture vote, lead to fewer instances of minority government which, in turn, tend to be less stable than majority cabinets. This would then suggest that the investiture requirement would correlate with higher levels of cabinet stability. In a somewhat different sense, Grof-man and Roozendaal (1997) point out that government formation processes that are particularly determinate will lead to more unstable cabinets than rules that create more uncertainty about the outcome of the government formation process. Specifically, government formation rules can reduce cabinet stability by allowing a party or parties "to expect to have no other chance to put together a cabinet" (440) which will in turn cause them to consider cabinet breakdown relatively less costly.
It is important to note that there are two streams of non-institutional explanations of cabinet stability in the literature (Grofman and Roozendaal, 1997). The first one is characterized by relating cabinet termination to critical events that take place outside the legislature and that may trigger a cabinet crisis by changing the preferences that actors have and the incentives that they face (Robertson 1983; Browne et al 1988, 1986; Frendreis 1986; Laver and Shepsle 1998). Such events j can be both political or economic, such as riots, rising levels of inflation, unemployment etc. However, there is no theoretical reason to expect a priori an event to become eventually critical in the sense of causing the premature termination of a government. Moreover, as Laver and Schofield (1990: 162) have noted, although events are important in affecting the durability of cabinets their effect is mediated through structures which make a particular cabinet potentially more or less resistant to a shock or crisis.
A second stream of non-institutional explanations (Grofman and Roozendaal 1994; Namd 1995) attribute cabinet durability to "the calculations made by rational actors in considering what the alternatives are to the continuation of the status quo and who has an incentive to prefer which of those alternatives" (Grofman and Roozendaal 1997: 442). Based on the assumptions that parties
The Institutional Sources of Cabinet Duration
107
care about controlling seats in the legislature; that parties value power within the governing coalition; and that parties can value some coalition partners more than others, Lupia and Strom (1995) develop a model that specifies the kinds of calculations that affect cabinet stability. They predict that a critical event will lead to the dissolution of the cabinet if and only if there is a legislative majority that prefers an election to leaving the governing coalition in power and all offering parties prefer an election to the best acceptable offer they can make to others to join the coalitions.
6.3 GOVERNMENT DURATION IN POST-COMMUNIST DEMOCRACIES
Before presenting the data on cabinet stability in the post-communist democracies, a couple of operational notes on the dependent variable are in order. The first pertains to the definition of what a cabinet is, i.e. when it starts and when it ends. Lijphart (1984: 265) notes that such a definition requires conceptual clarification about the event or events that mark(s) the end of one government and the beginning of a new one. He identifies seven such events in the literature: a new general election; any change in the partisan composition of the governing coalition; change in the parties that support the cabinet from the outside but do not formally participate in it; change in the coalitional status of the cabinet; change of prime minister for political reasons; change in prime minister tor any reason; and the resignation of the cabinet. After an empirical test of the different measurements using data from twenty West European democracies, Lijphart states his preference to define cabinet duration in terms of changes in the partisan composition of the governing coalition (1984: 278). In fact, he notes that nearly every work he has surveyed uses this event, in combination with others, in defining a cabinet. Budge and Keman (1990, 1993) have added change in the prime minister, a general election, and formal resignation to mark the beginning and the end of a cabinet. This definition has actually become the conventional norm in studies on cabinet government (Woldcndorp etal 1998, 2000) including the two extant surveys of post-communist cabinets (Blondel and Müller-Rommel eds. 2001; Müller-Rommel 2004).
The second operational question concerns the measurement of time for which the cabinet is in power. The key question here is not whether one uses months or days as the basic units of accounting for time, although the latter has become the norm. Inslead, a much more important decision to make is whether to count the actual length of time a cabinet lasted or the length of time a cabinet lasted as the fraction of the maximum length of time that it could have lasted (Lijphart 1984: 270). For instance, imagine a post-election government that is invested in
los
The Institutional Sources of Cabinet Duration
office immediately following a general election that lasted for exactly 365 days. Following the collapse of this government, a new cabinet is invested that lasted for die same length of time. In terms of the actual number of days both cabinets have lasted the same length of time. However, assuming a four-year legislative term, the duration of the first government was only one-quarter of the maximum time it could have lasted while the duration of the second government was exactly one third. As such as the second cabinet was relatively more stable even though in actual terms both cabinets lasted for the same length of time.
In the following overview of cabinet stability in the post-communist democracies, I adopt the conventional definition of cabinets as discussed above. That is to say that I shall consider any of the following changes as a change in the standing cabinet:
1) change in the partisan composition;
2) change in prime minister;
3) resignation of the government when followed by the re-formation of the same cabinet;
4) a general election.
As for the unit of time, cabinet duration is measured by the actual number of days. When a cabinet ends because of a general election, I take a date of the election as the end of the cabinet's tenure in office. The start date of each cabinet is the day when it was offcially invested in office either by appointment of the head of state or by a parliamnetary vote of investiture.
Table 6.1 shows the average duration of all cabinets in each of the ten states. The Table also distinguishes two types of governments; those that arc formed immediately after a general election and those that are formed later in the term. The last row of the Table shows that the overall duration of post-communist cabinets is very short: the average duration of the 106 cabinets is only 551 days with post-election cabinets lasting almost exactly twice as long as cabinets that are formed later in the term (790 versus 382 days). However, there is variation among the ten states. Specifically, it is worth noting that in three states (Czech Republic, Lithuania, and Romania), mid-term cabinets last a little longer on average than cabinets that are formed immediately after the general election. In the case of the Czech Republic the difference is small, however in the other two states the difference is 100 days or more.
Table 6.1 provides strong preliminary evidence in favor of the view that the institutional foundations of power dispersion matter for the relative duration of governments. As we move down from the first to the third group of states, i.e. as we move from lower to greater degrees of institutional power dispersion, average cabinet duration clearly drops. The last column of Table 6.1 shows that the average number of days a cabinet lasts in the first group is 701 followed by 509.08 in the second and 473.96 in die third group. Wc find the same order in the second
The Institutional Sources of Cabinet Duration 109 Table 6.1 Cabinet duration in post-communist democracies, 1990-2010
State
Post-election cabinets
Mid-term cabinets
Ail cabinets
I Most favorable institutional conditions for power concentration
Estonia	714.00	380.67	514.00
Hungary	980.40	530.25	780.33
Bulgaria	1114.20	284.33	803.00
Slovenia	820.25	666.75	743.50
Group average	937.66 (432.62)	451.41 (331.79)	701 (454.34)
1. Mixed institutional conditions for power concentration			
Latvia	466.20	288.00	337.50
Slovakia	1321.25	147.50	930.00
Group average	7X7.66 (532.13)	304.4(195.75)	509.08 (423)
3. Least favorable conditions for power concentration			
Lithuania	455.50	540.29	509.45
Poland	552.80	313.78	399.14
Romania	361.00	469.60	438.57
Czech Republic	604.25	631.75	618,00
Group average	635.82 (406.49)	382.23 (296.27)	473.96 (357.74)
N	44	62	106
Total	790.36	382.10	551.56 (416.22)
Mole: The numbers in brackets indicate standard deviation values.
column, which indicates the average number of days that post-election governments lasted. However, with respect to mid-term governments, this pattern is no longer as clear: while the most stable mid-term cabinets are still found in the first group, the average duration of mid-term cabinets is longer in the third than in the second group. The contrast between the first and third groups is particularly obvious and striking. In terms of post-election cabinets, each and every state in the first group has a higher value of duration than any state in the third group. Almost the same holds true with regard to the overall level of cabinet duration with the exception of the deviant cases of Estonia and the Czech Republic.
Minority governments in general are more durable than minimum winning coalitions as they last for an average of 537 days as opposed to the latter's 423 days. However, the most stable governments are those that are formed by a single majority party (820 days) followed by oversized coalitions (652 days). Controlling for the post-clcction status of the cabinet changes these relations somewhat. Single-party majority cabinets are still the most durable, whether they are formed immediately after an election or later in the term, but minority governments are the least stable type of cabinet in both groups. Of the two types of majority coalition cabinets, minimum winning coalitions are relatively more durable than oversized coalitions only when they are formed immediately after a general election.
These patterns vary across the ten states considerably partly because the four cabinet types are not present in each and every state. In fact, minority governments
1
I 10
The Institutional Sources of Cabinet Duration
are the only type of government that can be found in all ten states, followed by minimum winning coalitions which can be found at least once in all but one state (the exception being Romania), and oversized coalitions which were formed at least once in seven states (the exceptions are the Czech Republic, Estonia, and Poland). There are only two states (Bulgaria and Lithuania) where political parties have formed each of the four cabinet types at least once. In both of these states minimum winning coalitions are actually the most durable while minority cabinets are the least durable types of government. Thus, even though single-party majority governments appear to be the most stable type of government when we consider all 106 cabinets in all ten states pooled together, in the two states where they were actually formed they are not the most stable type. Furthermore, Bulgaria and Lithuania also differ with regard to the relative stability of their oversized coalitions and single-party majority cabinets.
Although minority governments are generally unstable, they are actually the most durable type of cabinet in two of the ten states (the Czech Republic and Romania) while everywhere else minority governments are the least durable cabinet type. There are similar variations with regard to the relative stability of minimum winning and oversized coalitions. Minimum winning coalitions are the most stable type of government in five states: Bulgaria, Estonia, Lithuania, Poland, and Slovenia, although the actual duration of these governments varies tremendously across the states. In the remaining three states (Hungary, Latvia, and Slovakia), oversized coalitions are the most stable.
6.4 AN INSTITUTIONALIST ACCOUNT OF GOVERNMENT DURATION
In this section I evaluate the effects that the four main institutions of power concentration/dispersion have on cabinet duration in the ten post-communist democracies. Specially, I will examine the following hypotheses:
Hypothesis la. District magnitude will have a negative effect on government duration: the lower the magnitude the longer lasting the cabinets will be. Democracies with more restrictive electoral systems will have fewer parties in their legislatures and, therefore, in their executive cabinets as well. With fewer parties, holding the level of their ideological dispersion constant, there will be fewer fault lines along which the government can be internally de-stabilized. Hypothesis lb. Party system fragmentation will inversely affect cabinet duration. The electoral system affects cabinet duration indirectly, i.e. via the political parties that make up the legislature and the executive. Electoral systems with lower district magnitude should have fewer parties, which in turn should result in
The Institutional Sources of Cabinet Duration
111
more durable cabinets. Conversely, electoral systems with larger magnitudes shouU lead to party proliferation and less stable cabinets. This hypothesis taps into the more direct connection between cabinet duration and a consequence of the electoral system, i.e. the number of parties.
hypothesis 2: Assembly size will inversely affect cabinet duration. For similar reasons as the electoral system, assembly size is expected to exert its impact on cabinet duration through the party system. Since larger assemblies are more likely to lead to greater degrees of party system fragmentation in the legislature than smaller assemblies do, I expect that larger assemblies will also reduce cabinet duration relative to smaller parliaments.
Hypothesis 3: Bicameralism will inversely affect cabinet duration. Bicameralism increases the number of veto points and veto players relative to unicameral legislatures, which, in turn, increase the coordination costs of keeping the government in office and in control of the policy and legislative agenda. It creates an additional institutional setting where the unity of the incumbent majority has to be tested and maintained. Relative to unicameral settings, this puts more stress and strain on the government and thus increases the likelihood of its premature termination.
Hypothesis 4a: Positive government formation rules will lead to more durable cabinets. Since positive investiture rules require the careful negotiation of a legislative majority that will support the government, it is more likely that the kinds of accommodations and concessions that might otherwise jeopardize the coalition mid-term would be addressed and taken care of before the government actually starts its term. In contrast, negative formation rules do not require such coordination and allow a government to enter office after simply having mustered an ad hoc support of partners on a single confidence vote.
Hypothesis 4b: Constructive no-confidence provisions should lead to more durable cabinets. This constitutional provision is normally entrenched with the precise intention of increasing cabinet duration by making it much more difficult to dismiss a government than what would be normally the case under ordinary no-confidcncc rules.
Hypothesis 4c: Majority governments will be more durable than minority cabinets. This hypothesis tests the relationship between cabinet status and duration directly. Majority cabinets are by definition winning coalitions, where every member receives positive power pay-offs as a reward for joining. In contrast, minority cabinets are not winning coalitions, although they may well be viable (Strom, Budge and Laver 1994), which in turn limits the degree and amount of power they can distribute to their members. All else equal, minority governments should lead power-seeking parties to search for alternative cabinet solutions more frequently relative to majority coalitions, resulting in higher levels of duration for the latter.
112 The Institutional Sources of Cabinet Duration
Hypothesis 5: Strong presidencies reduce cabinet duration. It is well known from the literature on the politics of presidential and semi-presidential systems that divided authority over the cabinet between the legislature and a directly elected president reduces cabinet duration (Shugart and Carey 1992). Maeda and Nishikawa (2006) report that governing parties in presidential systems face increasing hazard rates of the termination while those in parliamentary systems face declining hazard rates. According to them, this difference I attributed to the different types of goals that the two constitutional arrangements encourage political parties to seek: parties in presidential systems are primarily vote-seekers, which encourages oppositions parties to coordinate and stage a powerful challenge to the ruling party in the next election, while parties in parliamentary systems can balance the vote seeking imperative with the other two main goals of party behavior, i.e office- and policy-seeking. Although Maeda and Nishikawa do not speak to semi-presidential systems as such, the logic of their argument clearly implies that semi-presidential systems should also lead to less durable cabinets than parliamentary systems. In the context of post-communist states, Harfst (2000: 34) finds that parliamentary systems with an indirectly elected president have the longest lasting cabinets on average, followed by those with a directly elected president, premier-presidential and president-parliamentary systems.1
Presidential powers may also have an indirect effect on reducing cabinet stability through their negative impacts on the institutionalization (Mainwarring 1993), the strength (Shugart 1998), and the number of political parties (Mainwaring 1993; Clark and Wittrock 2005), all of which can, in turn, reduce the duration of cabinets. The impact of presidential powers on the number of parties is of particular significance: strong presidencies may encourage the fragmentation of the legislative party system by discouraging the formation of large parliamentary blocs, and the coalescence of actors behind large parties, given that both the policy-authority and the overall political power of the legislature are comparatively weak (Clark and Wittrock 2005: 175-6). Indeed, Maeda and Nishikawa also report that parliamentary systems tend to have less fragmented party systems than presidential ones (2006: 364).
Table 6.2 reports the results of a series bivariate OLS regression results. The dependent variable is the actual number of days a cabinet lasted in office. Most of the independent variables were defined in previous chapters except for the effective number of parties, which is measured by the Laakso-Taagepera (1979) index. The variable Low Magnitude is a dummy with a value of 1 for every election in which the average district magnitude is less than ten. Otherwise the value is 0. The results support most of the hypotheses, tire exception being Assembly size. Cabinet duration is a negative function of the number of parties, which is an important outcome of the electoral system, when regressed on the Effective number of parties. Both bicameralism and a strong presidency, institutions that disperse
The Institutional Sources of Cabinet Duration
le 6.2 Bivariate OLS regression results of institutional effects on cabinet duration in post-communist democracies
113
Variables	Coefficient	r
Low magnitude	142.792 (81.073)	0.019
Effective number of parties	-59.313 (24.065)**	0.0465
Large assembly	36.324 (81.908)	-0.008
Assembly size	-0,0954(0.301)	-0.008
Bicameralism	-136.541 (85.304)	0.0145
Positive investiture	61.796 (83.136)	-0.004
Constructive no-confidence	65.042 (89.364)	-0.004
Strong president	-172.498 (83.115)**	0.031
Majority government	332.817 (75.831)*	0.149
Post-election government	416.926 (72.533)*	0.235
N	106	
Note: * p<0.01; ** p<0.05. The numbers in brackets indicate standard		error.
political power, reduce cabinet duration relative to unicameralism and democracies with a weak presidency. Positive government formation and termination rules increase cabinet duration as does the cabinet's majority status. A dichotomous definition of assembly size (Large assembly) suggests that first chambers with fewer than 200 seats see more durable cabinets than those with fewer seats. However, measured by the actual number of seats in the first chamber results in a negative effect on the dependent variable. Finally, I also checked for the effect of post-election status on cabinet duration and found a strong statistically significant positive effect. In fact, of these bivariate relationships it is the last one that explains the highest percentage of variation in cabinet duration.
An important finding that emerges from Table 6.2 is that similarly to Western Europe the number of parties has a negative impact on cabinet duration in the post-communist democracies. In a survey of seventeen post-communist states that joined the Council of Europe, Harfst (2000) also finds that the fragmentation of the party system provides a more robust predictor of government duration than either the level of democracy or the type of executive-legislative relations enshrined in tire constitution. Of the seventeen states he finds seven deviant cases where the number of parties does not correlate with cabinet duration in the expected direction. Three of these states, Bulgaria, Lithuania, and Romania, are also included in our sample; in each of these cases Harfst reports that cabinets are less stable than expected given the relatively low level of fragmentation in their party systems.2 Similarly, Kluonis (2003 number of parties has a negative effect on cabinet duration (r same ten states that I am examining; in fact, the correlation is very similar to what Taylor and Herman (1971) had reported for West European democracies almost three decades earlier. In two separate publications on post-communist cabinet stability, Nikolenyi (2004, 2005) shows that post-communist legislatures with a
107) also reports that the effective 2 -   0.44) in the
s
^       cv t
^ q Oi f,
^' *d ^ * c «
oo -=fr m —■ o
|   rj rn «t —■
I ? I I I I I B I
~. < o o 2 —; r-i o
m oo I
t Q\ V f> > fN m c*i i CI A m
oo      o>, r-~ <-i-i c*)
O 0\ -h >B h c — — m m oo
r*l os oo .©
■a ■§ a
1 1 8
.Son
a I g v ra
ill! I
4- B
E 1
g a
77ie Institutional Sources of Cabinet Duration
115
dominant player tend to have more stable cabinets, especially when the dominant player also occupies the center of the party space, than legislatures with no j0Jflinant player present. By definition, the latter are more fragmented that the former, which suggests that Nikolenyi's findings also support the conventional wisdom about the relationship between party system format, i.e. fragmentation, government duration. In Table 6.3 I estimate the effects of all hypothesized institutional variables on cabinet duration. Model 1 presents the most complete equation with seven dichot-omously defined dummy variables. All but one variable points in the expected direction, however, only two of them have statistical significance: Majority and post-election status. The puzzling exception is the Constructive no-confidence variable, which is curious, because, as I mentioned earlier, this is the only institution that is entrenched in the constitutions with the express purpose of promoting cabinet stability. Models Two through Five replace Low magnitude with the Effective number of parties; the variable retains its negative sign consistently although it is not significant in any of the models. There are four additional variables that have a consistent impact on cabinet duration in all models: Bicameralism and Strong presidency have a negative, while Majority and Post-election cabinet status have positive impacts.
6.5 CONCLUSION
Government duration is at the very heart of party government and coalition politics. To the extent that governments last and work, policies will be formulated and implemented which will not only contribute to the collective welfare of society but will also create the conditions for retrospective voting and allow political parties, both in government and in opposition, to develop their identity and electoral following along programmatic and policy lines. On the other hand, if governments do not last, if coalition infighting leads to frequent inter-election cabinet turnovers, then neither parties nor voters will be in a position to develop a programmatic orientation that can serve as the foundation for the development of a well structured and stable party system. Instead, voters will be fickle, elections will be volatile, and parlies will have no incentive to remain unified and cohesive entities. In extreme circumstances, party/coalition government may undermine the viability of the new democracy if it adds too much stress to an already fragile and nascent framework of political competition. In short, government duration matters and its institutional underpinnings are particularly important to understand for those who seek to build and design well functioning newly democratizing polities.
This chapter has shown that institutional design has a strong impact on the duration of governments in the post-communist democracies. Specifically, the
116
The Institutional Sources of Cabinet Duration
central findings of this chapter are as follows: (i) systems with a weak presidency have longer lasting cabinets than those with a strong presidency; (ii) bicameralism reduces cabinet duration relative to unicameralism; (iii) cabinets formed under positive investiture rules last longer than those that are formed under negative parliamentary rules; (iv) party system fragmentation has a negative effect on cabinet duration; (v) post-election cabinets last longer than those formed midterm; and that (vi) majority governments are more durable than minority cabinets.
In sum, this chapter further confirms that the institutional design of power dispersion and power concentration matters for the unfolding pattern of coalition politics and party government in the post-communist democracies. As we saw earlier, political systems with institutions that concentrate political power tend to promote the formation of sizable electoral coalitions, followed by the formation of majority governments, which tend to last longer than minority cabinets, which are promoted by the institutional mix of power dispersion.
NOTES
It is worth noting that Harfst's sample includes non-democratic post-communist systems such as Moldova, in the premier-presidential category, Macedonia, in group of parliamentary systems with a directly elected head of state, and Albania, in the group of purely parliamentary systems. He rinds a mean cabinet duration of 633 days in pure parliamentary systems, 590 days in those where the president is directly elected and 479 days in the group of premier-presidential systems. It is worth noting that the three president-parliamentary systems covered by his study (Croatia, Russia, and Ukraine) have the shortest mean cabinet duration at 408 days.
The other deviant cases are Albania, Croatia, Ukraine, and Georgia. In the latter, a high level of fragmentation coincides with relatively stable cabinets.
Dividing the Executive?
Party Coalitions and Indirect Presidential Elections
A
7.1 INTRODUCTION
Although there is no shortage of studies on post-communist presidencies and presidents (Lucky 1994; McGregor 1994; Baylis 1996; Hcllmann 1996; Fry 1997; Beliaev 2006; Tavits 2009), a literature on the role that political parties play in the selection of the chief executive is noticeably absent. One reason for this lacuna is that most presidents in the East and Central European new democracies are elected directly by the voters, which means that presidential elections are normally examined as a matter of electoral studies. However, the head of state is selected indirectly by the legislature, or a special electoral college, in four states (Czech Republic, Estonia, Hungary, and Latvia). In these cases, political parties are the key players in deciding the outcome of the presidential selection game; in fact, the presidential contest becomes another round of renegotiating coalition agreements, under new rules, not only within the incumbent government but also with the opposition. Invariably, these rules are considerably different from those that govern the investiture of the cabinet and the prime minister and require that the winning candidate should secure a special majority. The objective of this chapter is to examine how the rules of indirect presidential selection in the four states shape the ability of the incumbent governing coalition to have its presidential candidate elected. In particular, based on the expectation that similarity in rules leads to similarities in outcomes, 1 will show that the more congruent the presidential and prime, ministerial selection processes are the greater the likelihood that the same coalition will capture both offices. If and when the presidential selection requires a special majority that is very different from what the selection of the prime minister required, then the outcome of the two processes should reflect this difference.
A clear understanding of when and why incumbent governing coalitions succeed or fail to capture the presidency is not only a theoretically interesting issue but it also has clear political and practical relevance. Tavits (2009) has shown that divided government accounts in a large degree for presidential activism both in new and in established democracies with a non-hereditary head of state. However, her analysis does not extend to uncovering the reasons why co-habitation or a split executive might emerge in the first place. Furthermore, whereas Samuels and