Presidents versus Parliaments

The Dynamics of Political Regime Shift in Croatia, Moldova, Mongolia and Turkey

by Paula Ganga, PhD Student, Geortown University

For other papers in this series click here

As authoritarian regimes all across the globe toppled and fell in what Samuel Huntington came to call the “third wave of democratization,” the theoretical relevance of the best form of government increased exponentially. In the context of an open and functioning political system, the rules of the game began to matter in ways they did not under autocracy. Following the lessons from transitions in the 1970s and 1980s, Juan Linz argued that when choosing a regime type, countries opting for a parliamentary system fared better than those which adopted a presidential one. The ensuing theoretical debate on the “perils” of presidentialism and the corresponding “virtues” of parliamentarism had a profound resonance in the early 1990s when the main proponents of these theoretical standpoints developed their research. The echoes of this debate were evident not only in the vast new literature that developed on topics of regime and institutional design, but also had an effect on the real-world choices made by political leaders in regions as diverse as Eastern Europe and Africa.

However, this literature focused from the start on analyzing the advantages and disadvantages of the two types of regimes as they appeared in the context of countries in Latin America in particular and the conclusions of this research were meant to serve as a cautionary tale to countries building new institutional infrastructures after the collapse of authoritarian regimes. Although the authors in this debate argued intensely over what particular characteristics of political regimes were more “conducive to stable democracy” (Linz 1990a, 52) and even went as far as to advocate for regime change, they never took the additional step of analyzing the cases of already established regimes that change their regime type without perturbing the functioning of the state. Linz and the others limited their recommendations to countries faced with the question of regime choice only once it was clear that the previous arrangement was no longer in place. By simply assuming a more or less clean slate in regime choice, the literature has not only overlooked the internal dynamics of regime change, but has also not sufficiently considered the determinants of regime choice.

In analyzing why a state would switch its regime type in the absence of any significant political crisis, the cases of Croatia, Moldova, and Mongolia in the early 2000s and Turkey since 2010 represent an excellent opportunity to engage with the theoretical debate between Linz and the others at the beginning of the 1990s and test it against the background of recent regime choices as well as new theoretical developments. This paper will argue that Linz’s contribution does not fully account for the effects of the process of political regime change itself. Moreover, this school of thought suffers from an important flaw: it eliminates agency from the process of regime choice. The present work argues that it is the political context and the leaders involved in decision-making processes that need to be the focus in the change of political regime. In each case this article will show how political leadership and the power balance between the most important actors in the system determined the final regime change.

This paper will thus start by examining the competing theoretical explanations of what makes presidential or parliamentary regimes preferable. The first of these approaches will be that of Juan Linz, the initiator of this debate. His pro-parliamentary perspective will be analyzed in conjunction with criticisms by Donald Horowitz, Scott Mainwaring, Matthew Shugart and others. Second, the paper will engage with the theoretical approach of hayo and Voigt which argue that the political context is the main driving force behind the institutional modifications. The second part of the paper will look at recent cases of regime modification, from presidential to parliamentary in the cases of Croatia, Moldova and Mongolia, and the proposed transition from parliamentary to presidential regime in Turkey. In each instance the analysis will focus on the balance of power among the elites and the institutional mechanisms through which the change occurred.


1. Presidentialism versus Parliamentarism: Theoretical Approaches

This article explores regime choices with reference to two basic theoretical types: presidentialism and parliamentarism. At the core of the distinction between presidential and parliamentary regimes is the degree of separation of powers: in parliamentary systems, the executive (whether collectively as a cabinet, or individually as a prime minister) depends on the confidence of the legislative majority for its survival in office, thus making the two branches closely intertwined. By contrast, in presidential systems, the executive remains in office even in the face of significant legislative opposition, as both branches enjoy their own electoral source of democratic legitimacy from the people (Sartori 1994, 106-108).

The middle ground between presidential and parliamentary systems was first identified by Maurice Duverger based on the example of the French Fifth Republic. According to Duverger, “a political regime is considered semi-presidential if the constitution which established it combines three elements: (1) the president of the republic is elected by universal suffrage; (2) he possesses quite considerable powers; (3) he has opposite him, however, a prime minister and ministers who possess executive and governmental power and can stay in office only if the parliament does not show its opposition to them…” (1980, 166).[1]


Juan Linz and the “perils” of presidentialism

In discussing the possibilities for democratic transitions in the states that emerged from Communism in the 1990s, political scientists have sought to draw lessons from earlier transitions in Latin America and Southern Europe (Linz and Stepan 1996). One such lesson was that parliamentary regimes are more conducive to the development of democracy than presidential regimes and that examples of long lasting democracies number far more parliamentary systems than presidential ones (Linz 1990a). Though Linz himself acknowledges that both regime types have created stable democracies across the world, he argues that there are shortcomings associated with presidentialism that make it a less suitable institutional choice for fledgling democracies. Based on the experience of Latin American countries, Linz claims that presidentialism holds within itself a constant paradox. On one hand, the system creates a strong executive legitimated by popular vote; on the other, the constitutions of these systems always reflect the suspicion that the president will use this legitimacy to personalize the power of his office (Linz 1990a, 54). Therefore, presidentialism carries with it a high degree of conflict; moreover at any time the elected leader can fall to the temptation of marginalizing his opponents and further concentrating power on the basis of the popular mandate.

This dimension of conflict is visible in the “perils” Linz famously associates with presidentialism. First, in presidential systems the president and assembly have competing claims to legitimacy. Both are popularly elected, and the origin and survival of each are independent from the other. Since both the president and the legislature “derive their power from the vote of the people in a free competition among well-defined alternatives, a conflict is always latent and sometimes likely to erupt dramatically; there is no democratic principle to resolve it” (1994, 7). Linz contends that parliamentarism obviates this problem because the executive is not independent of the assembly. If the majority of the assembly favors a change in policy direction, it can replace the government by exercising its no confidence vote.[2]

Second, Linz argues, the fixed term of the president’s office introduces a rigidity that is less favorable to democracy than the flexibility offered by parliamentary systems, where governments depend on the ongoing confidence of the assembly. By virtue of their greater ability to promote changes in the cabinet and government, parliamentary systems afford greater opportunities to resolve disputes (1990a, 55).

Third, presidentialism “operates according to the rule of ‘winner-take-all’—an arrangement that tends to make democratic politics a zero-sum game” (Linz 1990a, 56) and makes the presidency “the biggest political prize to be won” (Lijphart 1995a, 130). By contrast, in parliamentary systems “power-sharing and coalition-forming are fairly common, and incumbents are accordingly attentive to the demands and interests of even the smaller parties” (Linz 1990a, 56). In presidential systems direct popular election is likely to encourage presidents to believe that they need not undertake the tedious processes of constructing coalitions and making concessions to the opposition.

Fourth, the style of presidential politics is less propitious for democracy than the style of parliamentary politics. The sense of being the representative of the entire nation may lead the president to be intolerant of opposition. The absence in presidential systems of a monarch or other ceremonial head of state deprives them of a person in a position of authority that can exercise restraining power (1990a, 62).

Finally, Linz argues that political outsiders are more likely to win the chief executive office in presidential systems, with potentially destabilizing effects. Individuals elected by direct popular vote are less dependent on (and less beholden to) political parties (1994, 11) and are also more likely to govern in a populist, anti-institutionalist fashion (Linz and Stepan 1996, 188).

While Linz’s argument has been persuasive enough to be accepted to large extents by many political scientists including Arend Lijphart and Arturo Valenzuela as well as Adam Przeworski, Michael Alvarez, José Antonio Cheibub, and Fernando Limongi. For example, Arend Lijphart considers the majoritarian tendencies of presidentialism to be especially harmful in divided societies, and that “rigidity and immobilism are the most serious general weaknesses of presidentialism” (1991, 76), thus explaining the “strong scholarly consensus in favor of parliamentary government” (1991, 48). Additionally Valenzuela argues that “[t]he fundamental weakness of presidential government is the frequent failure of presidents to secure cooperative legislative majorities, a problem that is particularly evident in multiparty or fragmented two party systems, and which badly aggravates the natural rivalries between branches of government” (1998, 124).[3]

Przeworski et al. build on Linz’s contribution by asking whether or not the type of regime influences the endurance of democracy. In their statistical analysis the results show that “Linz is right about the durability of alternative institutional arrangements” (1996, 45). In their selected time period, 14 democracies (or 28% of the 50 cases) died under a parliamentary system, only one (12.5% of 8 cases) died under a mixed system, and 24 (52% of 46 cases) died under presidentialism. Among those democracies that died, the parliamentary systems lasted an average of eight years, while their presidential counterparts lasted nine. In these conditions the life expectancy of democracies under presidentialism is less than 20 years, while under parliamentarism that number is 71 years.

For Przeworski et al, “the survival of democracies does depend on their institutional systems. Parliamentary regimes last longer, much longer, than presidential ones. Majority-producing electoral institutions are conducive to the survival of presidential systems: presidential systems facing legislative deadlock are particularly brittle” (46). Both systems are vulnerable to poor economic performance, but presidential democracies are less likely to survive even when the economy flourishes than are parliamentary systems when the economy declines. Thus, they conclude that “[t]he evidence that parliamentary democracy survives longer and under a broader spectrum of conditions than presidential democracy seems incontrovertible” (46).

Among the first in trying to prevent Linz’s take on the advantages of regime type from becoming “conventional wisdom in political science,” Donald Horowitz points out to three major weaknesses in the criticism of presidentialism. First, Linz’s argument is based on “regionally skewed and highly selective sample of comparative experience” (1990, 74). If a sample set drawn primarily from Africa would be used to analyze parliamentarism, similar conclusions about the regime design as a source of crisis could be drawn.[4] Second, the “perils” of presidentialism are based on a “mechanistic, even caricatured, view of the presidency” (Horowitz 1990, 74). Mainwaring and Shugart also agree with this latter argument by pointing out that Linz makes the contrast between the two regime types starker than it actually is (1997, 451). Moreover, Shugart and Carey (1992) call the attention to the fact that the category of presidential regimes is not a homogeneous one: there are important differences from country to country.[5] Not accounting for these nuances makes Linz’s conclusions too general according to Horowitz and not at all responsive to different institutional arrangements within the presidential regimes (Horowitz 1990, 75).[6]

Third, Linz’s analysis also suffers from his incorrect assumptions about how presidents are chosen in presidential systems. In Horowitz’ view, by conflating presidential elections with ‘winner-take-all’ effects, Linz completely disregards other possible ways of electing presidents that could be more conducive to coalitions and to cooperation. For example, by making the requirement for the winning candidate to obtain votes in all the regions of the country, some of the divisiveness and conflict Linz associated with presidentialism is attenuated. Finally, Linz also seems to ignore “the functions that a separately elected president can perform for a divided society” (74). While Linz argues that in divided societies parliamentarism is more likely to help avert political crises,[7] Horowitz provides the examples of African countries to show how parliamentarism was in fact the cause in the exacerbation of ethnic violence (1990 and 2001).

Horowitz thus concludes that “Linz’s quarrel is not with the presidency, but with with two features that epitomize the Westminster version of democracy: first, plurality elections that produce a majority of seats by shutting out third party competitors; and second, adversary democracy, with its sharp divide between winners and losers, government and opposition” (79). The superficiality of Linz’s objections to presidentialism makes it easy “to turn his arguments around against parliamentary systems, at least where they produce coherent majorities and minorities… As a result, Linz’s thesis boils down to an argument not against the presidency but against plurality election, not in favor of parliamentary systems but in favor of parliamentary coalitions” (79).

Additional critiques of Linz’s position have been raised by Mainwaring and Shugart. Although they agree that “the issue of dual legitimacy is nettlesome in presidential systems … conflicting claims to legitimacy also exist in parliamentary systems. Conflicts sometimes arise between the lower and upper houses of a bicameral legislature, each claiming to exercise legitimate power” (1997, 451).[8]

For Mainwaring and Shugart presidentialism actually has some very important and converse advantages. The direct election of the chief executive gives the voters two electoral choices rather than one. Voters can support one party or candidate at the legislative level while backing another at the executive level (Mainwaring and Shugart 1997, 460-461). Additional advantages are accountability and identifiability. The more straightforward the connection between the choices made by the electorate at the ballot box and the expectations to which policymakers are held, the greater is the electoral accountability (462).[9]

The authors criticize Linz’s assertion that presidentialism induces more of a winner-takes-all approach to politics than does parliamentarism. In their opinion, the degree to which democracies promote winner-take-all rules actually depends on additional factors beyond the choice of regime type, such as the electoral and party system or the federal or unitary nature of the system (453). Accordingly, the effects of electoral and party systems have been analyzed from the point of view of both presidentialism and parliamentarism. For example, research by Mainwaring (1990 and 1993) and Stepan and Skach (1993) has suggested that democracy is less likely to survive when the political system combines presidentialism with a fragmented party system. For Przeworski et al. “[c]ombining presidentialism with a legislature where no single party has majority status is a kiss of death” (1996, 45). However, the opposite point can be made. That parliamentary systems with disciplined parties and a majority party offer the fewest checks on executive power, and hence promote a winner-take-all result, is an observation that has been made independently by Mainwaring and Shugart (1997), Horowitz (1990), and Lijphart (2006). These deleterious effects make Lijphart suggest that in order to avert them, “a combination of parliamentarism and proportional representation should be an especially attractive [option] to newly democratic and democratizing countries” (2006, 73). Valenzuela also argues that “[t]he logical option for countries with fragmented or multiparty systems where presidents can’t count on structuring majority coalitions is to move toward full-fledged parliamentarism” (1998, 137).

Valenzuela and Lijphart’s suggestions are not uncommon for the literature on regimes and institutions in the early 1990s. As the tumultuous events of 1989 brought a number of countries from the socialist camp into an institutional limbo, the need for new institutional arrangements to ensure the most equitable way of dividing power prompted political scientists to step beyond merely analyzing the relative benefits of the different types of regimes to actually suggesting specific features to the drafters of the region’s new constitutions. In this regard, Juan Linz himself implicitly advocates switching to parliamentary government in two extended versions of the article in Journal of Democracy (1990, 101-102, and 1994, 68-69). Likewise, Horowitz notes that “[t]hese are important arguments, because democratizing societies need to think, and think hard, about electoral systems that foster conciliation and governmental systems that in include rather than exclude” (1990, 79).[10]

Similarly Lijphart noted: “If forced—or begged—to give advice, it seems to me that the evidence does point in the direction of moderate proportional representation as the safer choice. It appears to give an at least slightly better governing ability without having to pay a heavy price in terms of minority representation” (1995b, 373). By contrast, Mainwaring and Shugart are “less than sanguine about the results of shifting to parliamentary government in countries with undisciplined parties… [because] switching to parliamentary government could exacerbate problems of governability and instability unless party and electoral legislation was simultaneously changed to promote greater discipline” (1997, 468).

The recommendations made by the participants to this debate as well as the general way in which they view the link between institutional arrangements and the outcome on the political scene places both the pro-presidential and the pro-parliamentary camps in the same tradition that underscores structure over agency. While Cheibub rightly points out that “[t]he question that dominated the research agenda fifteen or twenty years ago [was] whether institutions matter” he also argues that “they are not the whole story” (2007, 25).


Political context and agency in regime choice

Reflecting on the influence of political context G. Bingham Powell argued that

Constitutions are man-made designs. These designs reflect the constitution-makers’ values, their expectations of the various arrangements, their own often laboriously negotiated compromises. In so far as constitution-makers have tried to make their constitutions responsive to local needs and conditions, and to avoid negative consequences, associations between constitutional type and democratic performance may reflect those initial efforts, rather than reflect ongoing consequences of the incentives and costs created by the constitution itself. In so far as the constitution embodies values widely held in a society, both constitutional type and performance pattern may be products of political culture—the configuration of attitudes and beliefs by citizens and elites in a society—rather than one being a cause of the other” (1982, 229 added italics).[11]


This excerpt from Powell’s 1982 Contemporary Democracies: Participation, Stability and Violence underscores precisely the dimensions that this paper argues are at the root of recent regime changes: leaders, their values, “their expectations of the various arrangements, their own often laboriously negotiated arrangements”, “local needs and conditions.” This observation has also been corroborated by recent research into what determines change in a country’s form of government.

Covering the period from 1950 to 2003 and identifying 123 switches in form of government, Hayo and Voigt conclude that for their sample of 169 countries a switch is more likely to occur if the amended constitution is parliamentary rather than presidential, if the country was never a British or French colony, and if the country is located in the Middle East, North Africa, Sub-Saharan Africa, East Asia, Southeast Asia, or South Asia. Also, the data showed that countries characterized by a high degree of ethnic and religious factionalism are more likely to switch their form of government. A change in regime is less likely to occur in former colonial powers and countries for which natural resource exports represent a high share of GNP (2010, 285).

According to this research, the main factors influencing the likelihood of a change in the form of government are political and related to intermediate internal armed conflict, sectarian political participation, degree of democratization, and party competition, as well as distributional aspects relating to knowledge (285). Furthermore, additional research concluded that changes in form of government can be explained by characteristics of the political system, internal and external political conflicts, and political leaders, whereas economic and socio-demographic variables do not matter[12] (Hayo and Voigt 2011, 2). Additionally, those countries in which constitutional change is initiated by referenda tend to move toward more presidential systems, with those in which constitutional assemblies or conventions are employed tend to select more parliamentarian forms of government.

While Hayo and Voigt take an important step by looking inside the black box of regime change and uncovering the factors determining such change, their analysis encompasses cases of regime change from as early as 1950. Also, most of the changes on the spectrum from parliamentarism to presidentialism did not occur in the conditions of a stable switch from one regime to the other. However, the emphasis placed on the domestic political dimension helped to shape the analysis of the case studies in the next section.


2. The Dynamics of Regime Change

Writing in the context of the early 1990s, Juan Linz and the others argued for regime switch in the context of cataclysmic transformations without stopping to consider the dynamics of the process of change, essentially assuming that their advice would be applied where the previous political order collapsed. While this assumption seems highly limited today, at the time it was seen differently. Analyzing regime change since the 1950s, Hayo and Voigt (2010) concluded that among the most important factors in regime change were factionalism, government crises, and armed conflict. Yet the reality of political organization has changed. Although violent regime change continues to be present and although there is still an important number of failed and fragile states, the international community seems to have evolved to include more stability, as evidenced by several recent cases of regime change without prior regime collapse.[13]

In further exploring the role of political agents in determining the shape of regime change, the paper will look at the changes that took place in 2000 towards parliamentary regimes in Croatia, Moldova and Mongolia.[14] Turkey will serve as the counter example of attempted regime change toward presidentialism. In each case several issues will be analyzed: the context of the transformation, the role of political actors (both in power and in opposition), the mechanism of change, the new institutional framework, and the outcome of this process.[15]

Choosing Parliamentarism: the Case of Croatia

When the League of Croatian Communists scheduled competitive elections in mid-1990, Franjo Tudjman and the HDZ (Croatian Democratic Union) rode a wave of popular nationalism and anticommunism to power. Interpreting their electoral victory as a mandate to secede from the Yugoslav federation, they led Croatia to independence a year later and ignited a war with Serbia (the dominant power within the federation), that would last until 1995. The military conflict and the resulting strongly-charged atmosphere of nationalism defined Croatian politics for most of the 1990s and transformed Tudjman into the widely revered “father of the nation” (Ottaway and Maltz 2001, 376).[16]

Until 2000, the ruling HDZ kept reconfirming its dominant position in all parliamentary elections (to the Chamber of Representatives in 1992 and 1995 and to the Chamber of Counties in 1993 and 1997). Tudjman won both presidential elections by a wide margin: 56,7% in 1992 and 60,3% in 1997 (Grbesa 2004, 59). Even though the system was a semi-presidential one, lack of a strong opposition allowed Tudjman to centralize power in the presidential office by controlling the prime minister closely, pressuring legislators into towing the party line, and often ruling by quasi-legislative decree (Søberg 2007). The HDZ only started to lose its power in the late 1990s as the war ended and the population began making demands for improved economic performance. In the context of Tudjman’s death in 1999, the 2000 elections became the decisive moment for the future of Croatian politics (Bellamy 2001).

In early January 2000, a six-party coalition led by the Social Democratic Party (SDP) and the Croatian Social-Liberal Party (HSLS) won 95 of the 151 seats to HDZ’s 46 in elections for the crucial lower house of parliament and designated SDP leader Ivica Racan as prime minister. A month later, HDZ candidate Mate Granic—considered the favorite in the elections—won only 22 percent of the votes in the first round for the presidency and was eliminated. In the second round, Stipe Mesic, the candidate of the small Croatian People’s Party (HNS), unexpectedly defeated HSLS candidate Drazen Budisa (Jasic 2000).

As Croatia wanted to distance itself from the image of a problematic Balkan country and demonstrate its ability to be part of the wider European community, political changes in early 2000 reflected such social aspirations and translated them into concrete policy. As soon as they took power, the new moderate leaders launched an effective public diplomacy campaign that unambiguously demonstrated Croatia’s willingness to join Europe, to tackle the sensitive issue of cooperation with the Hague Tribunal, and to normalize relations with their neighbors, particularly Bosnia (Massari 2005, 265-266). The policy priorities of the coalition, signed in a pre-election agreement, were to abolish semi-presidentialism and strengthen the parliament, depoliticize the army and police forces, strengthen civilian control over the secret service, decentralize the state, revise suspect privatization cases, and pursue cases of corruption and abuse of power (Dolenec 2008, 39).

The constitutional amendments of 2000 were described as “the conversion of the Croatian political system from semi-presidentialism to parliamentarism” (Fish and Kroenig 2009, 173). Prior to 2000 the president had free reign to appoint the prime minister and the government, whereas after this moment he is required to make the appointment of a prime minister and government only if they enjoy majority support in the legislature. Since the amendments, the president has also lost the power to appoint five members of the Chamber of Counties (the upper house of the Parliament).[17] Additionally, the legislature gained the authority to influence the executive through interpellation, investigation, and votes of no confidence among others. Under the new provisions, the legislative did not share lawmaking powers with the president, who lacks decree, veto, and gate-keeping powers (Fish and Kroenig 2009, 174).[18]

In spite of the important advances, the road ahead was not entirely smooth for Croatia. The first coalition government (2000-2002) fell because of feuding among the key coalition partners; the second coalition government (2002-2003) lasted only a year until regular parliamentary elections were held (Kasapovic 2003, 52). The relatively modest achievements of the reforms introduced by the SDP-led governing coalition and the HDZ’s embrace of the goal of EU accession helped the latter party to win the election, although it was afterward forced to form a coalition government. Boosted by the EU’s decision in October 2005 to open negotiations over membership with Croatia, the HDZ narrowly held onto power in the 2007 elections (Pikering 2008, 534-535).

However, overall since 2000, Croatia has made immense progress in its transition toward democracy. Parliamentary elections at the end of 2003 and the manner in which the successive change of government was conducted, as well as the presidential elections in early 2005, indicate further democratization. Since changes were made to the government in 2000, democratic institutions in Croatia can be described as stable and capable of performing their functions, as evidenced by the consistently good scores the country has received from outside sources (see Table 2 for Freedom House scores) (Kusic 2006, 69) and as of 2013 Croatia has become a EU member state.

Choosing Parliamentarism: The case of Moldova

In the early 1990s Moldova was considered one of the most robust democratic polities in the former Soviet Union after the Baltic countries. In spite of “poverty, economic decline, ethnic conflict, and weak civil society and rule of law, Moldova … had come very close to meeting standard minimum definitions of democracy” (Way 2002, 130). This was particularly true in the way elections were contested in this young country. Parliamentary elections of 1990, 1994, 1998 and 2001 were highly competitive, even though the party system fluctuated dramatically from one legislature to the next.[19] This early period also saw the emergence of media sources presenting a wide range of views, even though many outlets were controlled or owned by the state or by political parties. Also, the Constitutional Court of Moldova acted independently to resolve disputes between the other branches of government. Furthermore, the legislature “consistently constrained presidential authority to a degree not seen in Moldova’s post-Soviet neighbors,” as strong parliaments are rare in the CIS (Way 2002, 130).

Between 1991 and 1994—that is, after independence but before the adoption of its new constitution—Moldova had been governed under the institutional framework it inherited from the Soviet Union. In 1991, Mircea Snegur, then a leading opposition figure, became Moldova’s first president. Still, the institution of a presidency was not created at the time. Rather, Snegur’s new “presidency” had been superimposed over the Soviet-era institutions that were left in place as the USSR was dissolving. This resulted in a constant power struggle between the executive and the legislative that could not easily be resolved (Mazo 2004, 12-13).

To a degree not witnessed in other post-Soviet states, the first legislature of Moldova was consistently able to constrain the president’s power. Yet when this constitution was drafted, a semi-presidential regime was chosen. This was a direct result of the political struggle between the two leading political figures of the country, Mircea Snegur and Petru Lucinschi. When the constitution was being drafted in 1994, President Snegur unsurprisingly preferred a stronger presidency, while Lucinschi, the head of parliament,[20] favored a semi-presidential model (Mungiu-Pippidi 2007).

The newly-created regime was a compromise semi-presidential arrangement which included both a president who served as head of state and a prime minister as head of government. However, in this institutional framework Moldova’s elected president would always be outmatched by the country’s legislature, which was consistently able to constrain his authority. Under the new constitution, the president could not circumvent the will of the legislature by resorting to referenda and only parliament could make amendments to the constitution (Matsuzato 2006, 329-330).

The 1998 parliamentary election ushered in an important change in the country: the re-legalized Communist Party (PCRM), which had been banned in 1991, managed to obtain 30% of the vote. After the redistribution of seats from wasted votes, the PCRM secured 40 of the 101 parliamentary seats (Mazo 2004, 28). This election also coincided with President Lucinschi losing the center-left parliamentary majority that he had enjoyed for the previous two years. During the next three years of uneasy “cohabitation,” five prime ministers were appointed and dismissed. This situation not only severely hurt the struggling Moldova economy but also its ability to secure foreign investment in the context of the regional consequences of the 1998 Russian financial collapse. Facing a deadlock, Lucinschi organized a non-binding referendum on the introduction of a more presidential system in May 1999 (Roper 2008, 120).

At the same time, a parliamentary alliance between the Communists and the center-right was taking shape. These two forces proposed amending the constitution to introduce a more parliamentary regime, with the president elected by the parliament. The amendments also eliminated the president’s ability to initiate legislation and transferred most presidential powers to the majority party or bloc in the legislature (Roper 2008, 120).

Contrary to what Moldova’s parliamentarians claimed at the time, the country did not become a pure parliamentary republic with the passage of these amendments. Instead, it now fit better in the “semi-parliamentary” category (Mazo 2004, 29). Although the president was now to be elected by parliament, he still served for a fixed term in office. He also retained the power to nominate the prime minister and was still the head of government. By retaining a fixed term after his parliamentary selection, the new president could not—unlike the prime minister—suddenly be brought down by a vote of no confidence. In short, the two branches, executive and legislative, still retained their fixed terms of office, even though they no longer had separate sources of popular legitimacy (Fish and Kroenig 2009, 451).

Even though Lucinschi’s term expired in the fall of 2000, he stayed on as acting president because no candidate was able to gather the required three-fifths majority in parliament. Centrists repeatedly blocked the election of the Communist leader, Vladimir Voronin. The ensuing gridlock eventually led to the parliament’s dissolution in January 2001 and to new elections a month later. These resulted in an overwhelming Communist majority, which then voted Voronin into office. Although the Communists won just over 50% of the vote, they benefited from an increase in the threshold (passed the previous year) from 4% to 6%. As a result, several center-right parties, which ironically had supported the raising of the threshold, failed to qualify for seats allowing the PCRM to control the legislature (Quinlan 2004).[21]

Once Voronin took office, the political situation stabilized dramatically. The referring of constitutional questions to the Constitutional Court, which had often resulted in the defeat of the president and/or the prime minister, decreased significantly after 2001. Some experts maintain that Moldovan politics stabilized under Voronin not because the new constitutional regime functioned well, but because he was the leader of the largest organized party in the country and had leverage by which to influence the parliament’s activities (Matsuzato 2006, 332).

The regime change, which brought an end to “pluralism by default,” did not promote democratic consolidation in Moldova. The opposite result was achieved instead; democracy soon became less consolidated and was now marked by its lack of pluralism. During Voronin’s two terms as president, Moldova’s Freedom House democratization rating (see Table 2) dropped from “non-consolidated democracy” to “semi-consolidated authoritarian” regime (Mungiu-Pippidi and Munteanu 2009, 137). In August 2009, four Moldovan parties – Liberal Democratic Party, Liberal Party, Democratic Party, and Our Moldova Alliance – created a governing coalition that pushed the Communists into opposition. The new acting president Mihai Ghimpu tried to adopt a new version of the Constitution of Moldova. Yet after the constitutional referendum failed in September 2010, Marian Lupu was elected as the Speaker of the Parliament. After the Alliance for European Integration lost a no confidence vote, the Pro-European Coalition was formed on May 30, 2013.

Choosing Parliamentarism: the case of Mongolia

Among the former Communist countries of Eurasia, Mongolia has been the recipient of much praise. Calling it the “inner Asian anomaly” M. Steven Fish (2001) argues that Mongolia’s consistent democratic record in the 1990s was an expression among others of a consolidated party system and the lack of a dominating political figure. Other researchers have looked for an answer to Mongolia’s success in its Buddhist tradition and the legacy of Genghis Khan (Sabloff 2002) or in the choice of a weak presidency rather than full presidentialism, lack of ethnic divides within the population and availability of foreign aid (Fritz 2002). However, even though in 2002 Fritz praised Mongolia for having “escaped the democratic erosion that has become a serious threat in more Western post-communist countries” (75), six years later she called Mongolia a “deviant democracy” (2008, 766). Although electoral democracy had become the “only game in town” by the late 1990s in this country (elections had been held regularly, leading to several alternations in power), this electoral democracy combined with substantial—if still imperfect—civil and political liberties had not yet generated effective accountability almost twenty years after the beginning of its democratic transition (2008, 778).

In the political arena, one of the most important acts for the new Mongolian democracy was the adoption of the 1992 Constitution, which was to establish a genuinely sovereign state that would represent Mongolia’s national interest (Munkh-Erdene 2010, 313). As a result, the Constitution endowed the country with a popularly elected, fixed-term president as “head of state and the embodiment of the unity of the people.” The president was granted “negotiating” and “designating” authority in selecting the prime minister and forming the cabinet. On the other hand, the State Great Khural (the parliament), “the supreme organ of state power,” was endowed with unlimited legislative power, as well as with the task of actually forming the cabinet after the parliamentary election (314).[22] Yet, the president’s power in the selection of the prime minister and the formation of the cabinet contained a far-reaching potential.

By assigning the president the authority to designate as prime minister the candidate nominated by the majority to the State Great Khural for appointment, the Constitution made the Parliament’s prerogative to appoint the prime minister dependent upon the president’s designation of the candidate. Moreover, the Constitution did not clearly spell out how the negotiation between the president and the parliamentary majority was to proceed (Barbayar 2003, 46). In particular, the Constitution failed to determine whose voice should take precedence in the case of a difference of opinion, thus equipping the president with the power to reject the majority’s nominee for prime minister, paralyzing the State Great Khural’s prerogative to form a cabinet. Therefore, the Constitution contained a source of political gridlock which quickly became a tool in political machinations (Munkh-Erdene 2010, 314).

In 1992, the Mongolian People’ Revolutionary Party (MPRP)[23] won by a landslide the first national parliamentary elections under Mongolia’s new Constitution (71 of 76 seats in the unicameral Parliament). This first free election essentially created a one-party state, allowing little outlet for opposition voices and little meaningful debate within the Parliament. This in turn resulted in strong protests culminating in the opposition-led hunger strike in the spring of 1995 demanding, among other things, an independent media free of government control (Severinghaus 2001, 61-62).

In 1996, the pendulum swung the other way. The newly formed Democratic Coalition won a surprising victory over the MPRP by taking 51 of the 76 seats, becoming the country’s first ever democratically based government. The Coalition’s four years in power, however, were characterized by revolving-door governments (four in four years), political infighting within the Coalition itself and a highly publicized corruption scandal. Helped by the fact that the Coalition was one seat short of a quorum, the MPRP successfully used every tactic in the legislative handbook to disrupt and stymie the implementation of the Coalition’s legislative agenda (Barbayar 2003, 47).

Thus the formation of new cabinets ran into political and constitutional obstacles raised by President Natsagiin Bagabandi—whose position was technically non-partisan but who emerged from the MPRP in the 1997 presidential elections—and the MPRP members in the State Great Khural (25 members out of 76). Four successive governments were rotated during 1998. Sparked by the intense debate over the vague formulations in the 1992 Constitution on whether an MP can hold a ministerial post at the same time, this fierce debate escalated further when President Bagabandi vetoed seven consecutive times the Democratic Coalition nominees for Prime Minister. Finally, in December 1999, the State Great Khural overwhelmingly passed the first amendments to the 1992 constitution (Finch 2002, 40-41).

The seven amendments dealt specifically with the procedures for nominating and approving candidates to the premiership and the formation of governments (Munkh-Erdene 2010, 315-328).[24] On 14 December 1999, the amendments were adopted by Parliament by 68 to zero, ignoring all the proposals initiated by the president and the Constitutional Court.[25] With the 1999–2000 constitutional amendments, Parliament successfully claimed the monopoly of legitimate authority to form the cabinet, effectively transforming the political system into a parliamentary one (Fish and Kroenig 2009, 456).

In the 2000 elections, the pendulum of power swung back yet again to the MPRP. This time, they took 72 of 76 seats (95%) in the Parliament, creating another one-party government, similar in numbers to that in 1992. However, because of the effects of the electoral system, the MPRP  was able to obtain such a dominant position while garnering only slightly more than 50% of the popular vote; clearly not the popular mandate suggested by the number of seats the party won (Severinghaus 2001, 62).

The change in regime, while taking place after long negotiations, did solve the competing claims between the president and parliament. Once the division of labor between these two branches of government was established, the contentiousness that had been characteristic of politics in the late 1990s began to subside. However, in the context of a strong parliament, and as the relative support of the two major parties became more equal throughout the decade, the need for creating coalitions—or to encourage the defection of an MP from the other party—kept the political scene in Ulaanbaatar competitive without affecting the regime’s stability (Table 2).

Choosing presidentialism? The case of Turkey

The first Ottoman-Turkish Constitution dates back to 1876, and since then Turkey has adopted four constitutions (1921, 1924, 1961 and 1982) (Ozbudun 2011). Of particular importance is the 1924 Constitution adopted under a National Assembly dominated by the Republican People’s Party organized around Ataturk. In this document, the Grand National Assembly was considered the sole representative of the nation and both legislative and executive powers were concentrated in the Assembly (Ozbudun, 2011). In practice, the Assembly could exercise its executive authority through the President of the Republic elected by this Assembly and a Council of Ministers appointed by the President. However throughout Turkey’s multiparty (1925-1946) and single-party (1946-1960) years, the leaders of the executive often reduced the Assembly to a subordinate institution within the state (Arslan 2005, 134-135). An interesting fact is that during the debates on the 1924 Constitution, the Assembly rejected a proposal to give the President the power to dissolve the Assembly. This happened while Ataturk was at the height of his power and prestige.

While the 1961 Constitution expanded civil liberties and social rights, it also demonstrated the military’s distrust in political parties. This document institutionalized judicial review and the complete independence of the judiciary, thus creating a checks and balances system in Turkish politics (Cremer 2011-2012, 297). The establishment of an advisory board called the National Security Council, composed of ministers and the highest chief officers in the Turkish army, which provided the military a forum to voice its perspective, especially in matters concerning national security.

The drafters of the 1982 Constitution aimed to create an executive unaccountable to the Grand National Assembly and therefore civilian politics, with the expectation that the President would be heavily influenced by the military. The 1982 Constitution accordingly provided the President with numerous powers, including the appointment of personnel in the judiciary and military, specifically in institutions including the Council of Higher Education, the Supreme Council of Judges and Public Prosecutors, the Constitutional Court, and the Turkish Aimed Forces. The changes implemented with the new constitution included the strengthening of the military’s presence in existing institutions, as well as in the creation of new institutions. Another critical change introduced by the 1982 Constitution, which is still debated today, involves the constitutional powers and immunities granted to the military and the National Security Council (Cremer, 2011-2012, 297-299).

Although modified several times in the last three decades, specifically within the framework of European Union reforms (Kalaycioglu 2011), the 1982 Constitution has allowed the military to remain highly influential in Turkish politics. The most significant reforms took place in 2010, when the ruling Justice and Development Party (Adalet ve Kalkinma Partisi – AKP) announced a constitutional reform package, aimed at democratizing the 1982 Constitution (Cremer 2011-2012).

The AKP’s rise in power dates back to the elections of 3 November 2002, when the three governing parties that had formed the coalition government after the 1999 national elections, as well as the two opposition parties, fell short of the 10 percent threshold and did not enter the parliament. Helped by the same threshold, the AKP won 363 seats with just 34.2 percent of the vote and formed a strong majority government. The AKP had only recently come into existence as one of two parties growing out of the banned Virtue Party, whose Islamic identity and discourse had caused the Constitutional Court to rule it a threat to the secular foundation of the Turkish Republic. Since 2002, the AKP has strengthened their political position due to significant political accomplishments. Turkey has concluded its accession partnership with the EU (albeit without actually becoming a full member), adopted a series of extensive political-reform packages, restructured civil-military relations to make the military less powerful, taken significant steps toward addressing the question of the Kurds, and followed a proactive foreign policy (Muftuler-Bac and Keyman, 2012). The economy has grown rapidly though the economic success has been brought into question by some economists (Rodrik 2013).

Due to the Islamist ideological background of many AKP members, the 2010 constitutional reforms caused turmoil among some citizens who perceived the proposal as a threat to Ataturk’s secular democracy, and they demanded Turkish military intervention. Despite the ideological turmoil, the Turkish public voted to amend the Constitution on September 12, 2010—exactly thirty years after the military intervention (Cremer 2011-2012).

Thus a large part of Turkish society, longing for political stability and effective governance above all, welcomed the AKP notwithstanding its origins. The AKP won again in July 2007, this time with an unprecedented 47 percent of the vote. Local elections in 2004 and 2009 yielded the similar results. The military, the judiciary, the media, opposition parties, and civil society organizations have tried through various means to curtail the AKP, generally claiming that it threatens Kemalist secularism, but none of these efforts have succeeded. In 2008, for instance, the Constitutional Court declined a state prosecutor’s request to close down the AKP on such grounds (Muftuler-Bac and Keyman, 2012).

Additionally, the tensions within the society and the political system only increased; the political environment preceding the 2011 Turkish elections was very hostile and characterized by intense debates among the AKP and the remaining three major political parties (Kubicek 2011). Apart from widespread accusations about corruption, indecency and others among the political parties, there were extensive protests and violence in some parts of the country. Moreover, there was ample evidence of the Prime Minister Recep Tayyip Erdogan’s strong desire to restructure the central political institutions along the lines of France’s semipresidential system (Cengiz and Hoffman, 2012).

In November 2012, Erdogan’s AKP officially announced its goal of introducing a “Turkified version of the U.S. executive system.” This was not the first time the AKP has advocated a strengthened executive branch; the office of the presidency was previously bolstered by a 2007 amendment declaring the president would be elected by popular vote and have slightly more power to influence the legislative process (Krueger and Chugh 2013). This system was appropriate to be implemented in the Fifth Republic well and there is no reason why it could not also be applied successfully in other countries. Yet, the French system was brainchild of General de Gaulle in a time of crisis when the French Forth Republic’s fragile political system that had let to an unstable and often dead-locked parliament with ever-changing governments and shifting parliamentary majorities (Gaffney, 2010). The situation in Turkey is not comparable to the France of late 1950s. There is no evidence of institutional inefficiency in Turkey’s current parliamentary democracy that would justify such a drastic systemic change. On the other hand, in the light of the tendencies for centralization of power (Benhabib 2013), there is ample evidence of the need for stronger constitutional protection of fundamental rights and freedoms in this country (Cengiz and Hoffman 2012, 266).

The post-2013 presidency that Erdoğan envisions will be substantially more powerful and independent than the 2007 amendment dictated. Rather than playing a primarily ceremonial role, this new proposal would give the president power to dissolve Parliament and appoint and dismiss ministers, ambassadors, and senior officials without parliamentary approval. The AKP and others in favor of the proposal argue that this system will be more stable and efficient, ridding Turkey of the fragile coalitions and political bickering that have plagued its parliamentary system in the past.  Critics, however, see the plan as a ‘power-grab’ by Erdogan, who intends to run for president if it passes[26] (Krueger and Chugh 2013).

The year 2013 has been dramatic in terms of developments in Turkish domestic politics. In March, the government of Prime Minister Recep Tayyip Erdogan’s Justice and Development Party (AKP) brokered a historic ceasefire between Kurdistan Workers Party (PKK) militants and the Turkish military, marking a critical first step in the reconciliation process between Turkish Kurds and the state. Subsequently, on May 28th a series of anti-government protests broke out, originating in Istanbul’s Taksim Square and quickly spreading to other parts of Turkey. Before the Taksim Square protests thrust Turkey into the international spotlight, the constitutional reform process had reached a stalemate within the government’s Constitution Reconciliation Commission (CRC). While the protests are unlikely to directly challenge the constitutional deadlock, they threaten to erode Prime Minister Erdoğan’s legitimacy and weaken the chances of adopting the presidential system that his party advocates (Krueger and Chugh 2013).

The likelihood of adopting presidentialism in Turkey is intrinsically linked to the fate of Erdogan’s AKP— a subject of much attention since the Taksim Square protests began. The protests may an adverse effect on Erdogan’s reputation, yet he still remains an exceptionally popular leader.  The anti-government protesters in Taksim Square and other cities across Turkey did not include the critical demographic of primarily rural, conservative citizens who support the AKP. A recent Pew poll reported a 62 percent positive approval rating for the Prime Minister.  Another survey conducted by Turkey’s MetroPOLL indicated an 11 percent drop in AKP popularity after the protests (Krueger and Chugh 2013). These numbers suggest that the fallout from recent protests is unlikely to force Erdogan from power. Nevertheless, Erdogan and the AKP’s popularity have certainly declined since the last electoral cycle, before the protests began.

One of the more interesting confrontations might take place inside the AKP itself. If Erdogan planned to run for president unopposed, the opinion polls show that current President Abdullah Gul received wide approval for his more diplomatic reactions to the street protests compared to Erdogan’s. Gul may use his political clout decisively against the AKP’s plan for a presidential system, which he has already expressed doubt about to the Turkish press (Butler 2012).

Analysts have argued that low popularity could force Erdogan to seek out another option allowing him to remain in power, such as extending his prime ministerial term limit, which is imposed by AKP internal party rules rather than the constitution (the limit is three terms as Prime Minister which Erdogan has already had) (Butler 2012).

The case of Turkey’s current process of transition from a parliamentary to a presidential regime perfectly encapsulates the “perils” Juan Linz associated with presidentialism. The possibility that the president will view the popular vote as a plebiscitarian means to impose his or her policy agenda without searching for a wider political support has already been signaled by Turkish analysts as part of Erdogan’s view of politics. Critics has argued that Erdogan has a “majoritarian view of democracy” (Benhabib 2013, The Economist 2013, Arato 2010) which is in line with Linz’s argument of the peril that the president—in this case prime-minister—will use the ballot as the supreme and final validation of their power.


Regime change: an ongoing debate

The search for the most ideal system of government has been a challenge for intellectuals and scholars since even before the foundation of the modern discipline of political science. To Rousseau’s formulation of this age-old question in his Constitution of Poland (“Which institutions have which effects under which historical conditions?”), Przeworski answered that “democracies are not all the same. Systems of representation, arrangements for the division and supervision of powers, and methods of organizing interests … can and do vary widely among regimes that are generally recognized as democratic” (1996, 53). This observation casts doubt on the claim that one type of regime—whether presidential or parliamentary—is better (or worse) than others.

This paper has carefully considered the early 1990s theoretical debates on ideal regime types, acknowledging their contribution in pointing out the relative benefits of each system. Engaging both with the contributions of Linz and his intellectual disciples, as well as the theoretical contributions on the economic benefits of different constitutional arrangements, this paper explored an area not addressed by previous research. The phenomenon investigated in this paper—in which countries switch from one regime type to another while averting regime collapse and remaining within a broad framework of constitutionally-established norms—has not received much previous scholarly attention. Exploring the cases of regime change in Croatia, Moldova and Mongolia as well as the recent debate around Turkey’s constitutional reform offered a look inside the processes that lead to the “perils” and “virtues” of regime types, while also detailing the dynamic effects of the changes themselves. Opening the black box of the actual process of regime change helps to illustrate the effect of agency in shaping outcomes previously analyzed as solely resulting from the institutional structure.

The argument of this paper on the importance of agency was supported by the findings in the three case studies. Political leadership was the main force behind those regime changes. Whether it was for the purpose of leaving behind an authoritarian past in Croatia, preventing the rise of a strong presidency in Moldova, clarifying the division of power between the branches of government in Mongolia, or enshrining a particular political context for the benefit of a leader in Turkey, the benefits and the disadvantages of regime types acquire entirely new dimensions once they are recast in the context of concrete political struggles.


Table 1. Summary of the case studies








Prior Context

Transition after the death of the previous leader

Deadlock due to “cohabitation” between president and the parliamentary majority

Deadlock due to unclear separation of powers

No apparent crisis in the constitutional functioning of the system, but protests on AKP continuing rule

Political actors

(support and protest to the regime change)

A broad coalition that won the elections. The HDZ had to accept the change due to electoral defeat.

Amendments as a way to prevent the president from changing the regime himself. A large parliamentary majority against president voted on the change.

Both parties in power and in opposition wanted to end the deadlock especially as new elections were approaching and neither wanted to govern in the same conditions.

Mainly AKP as the main governing party under the leadership of Recep Tayyip Erdogan

Mechanism of change (referendum/amendments)




Constitutional Reform

New regime type






Slight instability at first due to the broad coalition in power with increased progress towards EU membership.

The change coincided with a return to power of the Communists for two mandates which affected the country negatively.

The clearer separation of powers ended the competition between the legislative and the president.



Table 2. Freedom House scores for Croatia, Moldova, Mongolia. The scores represent Freedom House index for political rights (1=most free; 7=least free) and civil liberties (1=most free; 7=least free) (Freedom House, Søberg 2007, 33; Mungiu-Pippidi 2007, 87)



























































Arato, Andrew. 2010. “The Constitutional Reform Proposal of the Turkish Government: The Return of Majority Imposition.” Constellations. 17 (2): 345- 350.

Arslan, D. Ali. “The evaluation of parliamentary democracy in Turkey and Turkish political elites.” Historia Actual Online 6 (2010): 131-141.

Barbayar, Tsedendamba. 2003. “Foreign policy and domestic reform in Mongolia” Central Asian Survey 22 (1): 45-59.

Bellamy, Alex J. 2001. “Croatia After Tudjman.” Problems Of Post-Communism 48 (5): 18-31.

Benhabib, Seyla. 2013. Turkey’s Authoritarian Turn, The New York Times, June 4, p. A5.

Butler, Daren. 2012. Turkey’s Erdogan sees tough path to constitutional reform. November 2, 2012, Reuters,

Cengiz, Firat and Lars Hoffman. 2012. The 2011 General Elections in Turkey: Potential Implications on Domestic and International Politics in the Shadow of a Discourse Change?, Parliamentary Affairs. 65: 255-269.

Cheibub, José Antonio. 2007. Presidentialism, parliamentarism, and democracy. New York, NY: Cambridge University Press.

Cremer, A. Serra. 2011. “Turkey between the Ottoman Empire and the European Union: Shifting Political Authority through Constitutional Reform.” Fordham International Law Journal 35: 279.

Dolenec, Danijela. 2008. “Europeanization as a Democratising Force in Post-communist Europe: Croatia in Comparative Perspective” Politicka Misao: Croatian Political Science Review 45 (5): 23-46.

Duverger, Maurice. 1980. “A New Political System Model: Semi-Presidential Government” European Journal of Political Research 8: 165–187.

Economist. 2013. “Democrat or Sultan. Recep Tayyip Erdogan should heed Turkey’s protesters, not dismiss them.” June 8.

Finch, Christopher M. 2002. “Political Consolidation and Continued Economic Reform” Asian Survey 42 (1): 39-45.

Fish, M. Steven. 2001. “The Inner Asian anomaly: Mongolia’s democratization in comparative perspective” Communist & Post-Communist Studies 34 (3): 323-338.

Fish, M. Steven, and Matthew Kroenig. 2009. The handbook of national legislatures: a global survey. Cambridge: Cambridge University Press.

Fritz, Verena. 2002. “Mongolia: Dependent Democratization” Journal Of Communist Studies & Transition Politics 18 (4): 75-100.

―—― 2008. “Mongolia: The Rise and Travails of a Deviant Democracy” Democratization 15 (4): 766-788.

Gaffney, J. 2010. Political Leadership in France: From Charles de Gaulle to Nicolas Sarkozy. Chippenham, Palgrave Macmillan.

Grbesa, Marijana. 2004. “Personalization in Croatian Presidential Election in 2000: How Personal Did the Candidates Go and What Did the Press Cover?” Politicka Misao: Croatian Political Science Review 41 (5): 52-73.

Hayo, Bernd and Stefan Voigt. 2010. “Determinants of constitutional change: why do countries change their form of government?” Journal of Comparative Economics 38 (3): 283–305.

―—― 2011. “Endogenous constitutions. Politics and politicians matter, economic outcomes don’t” Journal of Economic Behavior and Organization (article in press, online 1 October).

Horowitz, Donald L. 1990. “Comparing Democratic Systems” Journal of Democracy 1 (4): 73-79.

―—― 2001. The deadly ethnic riot. Berkeley: University of California Press.

Ilisin, Vlasta. 2007. “The Social Structure of the Croatian Parliament in Five Mandates.” Politicka Misao: Croatian Political Science Review 44 (5): 45-70.

Jasic, Suzana. 2000. “Monitoring the Vote in Croatia” Journal Of Democracy 11 (4): 159-168.

Kalaycıoğlu, Ersin (2011) The Turkish–EU Odyssey and Political Regime Change in Turkey, South European Society and Politics, 16(2): 265-278.

Kasapovic, Mirjana. 2003. “Coalition Governments in Croatia: First Experience 2000-2003” Politicka Misao: Croatian Political Science Review 40 (5): 52-67.

Kimura, Ehito. 2010. “Changing the rules: Historical conjuncture and transition in Indonesia” Asia Pacific Viewpoint 51 (3): 248-261.

Krueger, Katherine and Ayesha Chugh. 2013. Constitutional Reform in Turkey and the Recent Protests. ElectionGuide Digest (June 21).

Kubicek, Paul. 2011. “The 2011 Elections and Prospects for Change in Turkey.” Mediterranean Politics 16, no. 3: 443-448.

Kusic, Sinisa. 2006. “Croatia: Advancing Political and Economic Transformation” Journal Of Southeast European & Black Sea Studies 6 (1): 65-81.

Lijphart, Arend. 1991. “Presidentialism and Majoritarian Democracy: Theoretical Observations,” in Gyorgy Szoboszlai (ed.) Democracy and Political Transformation: Theories and East-Central European Realities. Budapest: Hungarian Political Science Association.

Lijphart, Arend, in collaboration with Don Aitkin. 1995a. Electoral systems and party systems: a study of twenty-seven democracies, 1945-1990. New York: Oxford University Press.

―—― 1995b. “The Virtues of Parliamentarism: But Which Kind of Parliamentarism?” in Chehabi, H. E., Alfred C. Stepan, and Juan J. Linz. Politics, society, and democracy. Comparative studies. Boulder: Westview Press.

―—― 2006. “Constitutional Choices for New Democracies” in Larry Diamond and Marc F. Plattner. Electoral systems and democracy. Baltimore: Johns Hopkins University Press.

Linz, Juan J., Arend Lijphart, Arturo Valenzuela, and Oscar Godoy Arcaya. 1990. Hacia una democracia moderna: la opción parlamentaria. Santiago, Chile: Ediciones Universidad Católica de Chile.

Linz, Juan. 1990a. “The Perils of Presidentialism” Journal of Democracy 1 (1): 51-69.

―—― 1990b. “The Virtues of Parliamentarism” Journal of Democracy 1 (4): 84-91.

―—― 1994. “Presidential or Parliamentary Democracy: Does it Make a Difference?” in Juan Linz and Arturo Valenzuela. The failure of presidential democracy. Baltimore: Johns Hopkins University Press.

Linz, Juan J., and Alfred C. Stepan. 1996. Problems of democratic transition and consolidation: southern Europe, South America, and post-communist Europe. Baltimore: Johns Hopkins University Press.

Lipset, Seymour Martin. 1990. “The Centrality of Political Culture” Journal of Democracy 1 (4): 80-83.

Mainwaring, Scott. 1990. “Presidentialism in Latin America” Latin American Research Review 25: 157-179.

―—― 1993. “Presidentialism, Multipartism, and Democracy: The Difficult Combination” Comparative Political Studies 26: 198-228.

Mainwaring, Scott and Matthew Shugart. 1997. “Juan Linz, Presidentialism, and Democracy” Comparative Politics 29 (4): 449-471.

March, Luke. 2004. “Socialism with Unclear Characteristics: The Moldovan Communists in Government” Demokratizatsiya 12 (4): 507-524.

Massari, Maurizio. 2005. “Do All Roads Lead to Brussels? Analysis of the Different Trajectories of Croatia, Serbia-Montenegro and Bosnia-Herzegovina” Cambridge Review Of International Affairs 18 (2): 259-273.

Matsuzato, Kimitaka. 2006. “Differing Dynamics of Semi-presidentialism across Euro/Eurasian Borders: Ukraine, Lithuania, Poland, Moldova, and Armenia” Demokratizatsiya 14 (3): 317-345.

Mazo, Eugene. 2004. Post-Communist Paradox: How the Rise of Parliamentarism Coincided with the Demise of Pluralism in Moldova. Center on Democracy, Development and the Rule of Law, Stanford Institute for International Studies, No. 17.

Muftuler-Bac, Meltem and Fuat Keyman, Turkey Under the AKP. The Era of Dominant-Party Politics, Journal of Democracy, 23: 1, 85-99.

Mungiu-Pippidi, Alina. 2007. “Disputed Identity as Unescapable Pluralism. Moldova’s Ambiguous Transition” Romanian Journal of Political Science 7 (2): 85-101.

Mungiu-Pippidi, Alina, and Igor Munteanu. 2009. “Moldova’s “Twitter Revolution”” Journal of Democracy 20 (3): 136-142.

Munkh-Erdene, Lhamsuren. 2010. “The Transformation of Mongolia’s Political System: From Semi-parliamentary to Parliamentary?” Asian Survey 50 (2): 311-345.

Ottaway, Marina, and Gideon Maltz. 2001. “Croatia’s Second Transition and the International Community” Current History 100 (649): 375-380.

Ozbudun, Ergun. 2011. The Constitutional System of Turkey: 1876 to the Present. Palgrave Macmillan.

Pickering, Paula M., and Mark Baskin. 2008. “What is to be done? Succession from the League of Communists of Croatia” Communist & Post-Communist Studies 41 (4): 521-540.

Powell, G. Bingham Jr. 1992. “Contemporary Democracies: Participation, Stability, and Violence,” in Arend Lijphart. Parliamentary versus presidential government. Oxford: Oxford University Press.

Przeworski, Adam, Michael Alvarez, José Antonio Cheibub, and Fernando Limongi. 1996. “What Makes Democracies Endure?” The Journal of Democracy 7 (1): 39-55.

Quinlan, Paul. 2004. “Back to the Future: An Overview of Moldova under Voronin” Demokratizatsiya 12 (4): 485-504.

Rodrik, Dani. 2013. Turkey’s Protests Have Been Let Down by All Sides, June 4, The Financial Times.

Roper, Steven D. 2008. “From Semi-presidentialism to Parliamentarism: Regime Change and Presidential Power in Moldova” Europe-Asia Studies 60 (1): 113-126.

Rossabi, Morris. 2009. “Mongolia: Transmogrification of a Communist Party” Pacific Affairs 82 (2): 231-250.

Sabloff, Paula L.W. 2002. “Why Mongolia? The political culture of an emerging democracy” Central Asian Survey 21 (1): 19-36.

Sartori, Giovanni. 1994. “Neither Presidentialism nor Parliamentarism,” in Juan Linz and Arturo Valenzuela. The failure of presidential democracy. Baltimore: Johns Hopkins University Press.

Severinghaus, Sheldon R. 2001. “Mongolia in 2000” Asian Survey 41 (1): 61-70.

Shugart, Matthew and John Carey. 1992. Presidents and Assemblies: constitutional design and electoral dynamics. Cambridge: Cambridge University Press.

Shugart, Matthew and Scott Mainwaring. 1997. “Presidentialism and Democracy in Latin America: Rethinking the Terms of the Debate” in Scott Mainwaring and Matthew Shugart. Presidentialism and Democracy in Latin America. Cambridge: Cambridge University Press.

Søberg, Marius. 2007. “Croatia Since 1989: The HDZ and the politics of Transition” in Sabrina Ramet and Davorka Matić (eds.) Democratic transition in Croatia: value transformation, education and media. College Station: Texas A&M University Press.

Stepan, Alfred and Cindy Skach. 1993. “Meta-Institutional Frameworks and Democratic Consolidation” World Politics 46: 1-22.

Valenzuela, Arturo. 1998. “The Crisis of Presidentialism in Latin America,” in Scott Mainwaring and Arturo Valenzuela. Politics, society, and democracy. Latin America. Boulder, CO: Westview Press.

Way, Lucan A. 2002. “Pluralism by Default in Moldova” Journal of Democracy 13 (4): 127-141.

―—―  2003. “Weak States and Pluralism: The Case of Moldova” East European Politics and Societies 17 (3): 454-482.

[1] Duverger’s conceptualization has been disputed by Shugart and Carey who argue that the term is misleading since the prefix ‘semi’ implies that these regimes are “located midway along some continuum running from presidential to parliamentary” (1992, 23). Instead they prefer the use of the term “premier-presidentialism.”

[2] While Linz makes this process seem very straightforward, a no confidence vote can also have very disruptive effects on the smooth functioning of politics in parliamentary system.

[3] Expanding on this point, he adds that “Presidentialism simply does not provide potent political incentives for coalition building and cooperation, even among members of the same party based in different branches of government.” In Latin America he observed a sharp increase in presidential prerogatives and the use of executive decrees. “Political crises that a parliamentary regime would have resolved via the replacement of a prime minister, the restructuring of a new working majority, or a calling of new elections, all too often tended to become full-blown constitutional crises under the conditions of presidentialism. Crises of government soon turned into crises of regime often with military intervention” (Valenzuela 1998, 130).

[4] To this critique Linz replied that in fact most of the world’s pure presidential systems with only a few exceptions were to be found in Latin America (1990b, 87).

[5] This is especially true when presidentialism in the United States is analyzed together with cases of presidential systems from other continents.

[6] Additionally, not only does presidentialism have several versions, but so does parliamentarism. In analyzing the “virtues” of parliamentarism Arend Lijphart is right to ask “which kind of parliamentarism” encapsulates these supposed advantages according to Linz (1995b).

[7] Linz’s opinion is supported by the statistical results of Przeworski et al. who argue that “only parliamentary systems are sensitive to the ethnic fragmentation of the population” making “presidential democracies simply more brittle” (46).

[8] They also argue that an additional potential source of conflicting legitimacy stems from the role of the head of state (even a ceremonial one), who is still called president in spite of being elected by parliament. “The constitutions of parliamentary republics usually give the president several powers that are—or may be, subject to constitutional interpretation.” Their examples include the president’s exclusive discretion to dissolve parliament and the requirement of countersignatures of cabinet decrees in Italy, suspensory veto over legislation and appointments to some high offices including ministries in Czech Republic and Slovakia (451).

[9] One objection to presidentialism’s claim to superior electoral accountability is that in most presidential systems one may not be reelected as head of state immediately, if at all. The electoral incentive for the president to remain responsive to voters is weakened in these countries, and electoral accountability suffers. In multiparty parliamentary systems, even if a citizen has a clear notion of which parties should be held responsible for the shortcomings of a government, it is often not clear whether voting for a certain party will increase the likelihood of excluding a party from the governing coalition (Shugart and Mainwaring 1997, 34).

[10] This point is taken further by his suggestion that “[p]rominent among innovations they might consider are presidents chosen by an electoral formula that maximizes the accommodation of contending political forces. Democratic innovators can only be aided by Linz’s emphasis on institutional design. But they can only be distracted by his construction of an unfounded dichotomy between two systems, divorced from the electoral and other governmental institutions in which they operate” (1990, 79).

[11] While the more general argument that political culture is an important dimension in studying the effects of regime type has been made by Seymour Martin Lipset as a direct answer to Juan Linz’s article “Perils of Presidentialism” (1990) the link between regime type and political culture is beyond the scope of this paper.

[12] Addressing the significant differences reported by Persson and Tabellini in terms of economic outcomes between presidential and parliamentary systems, the authors suggest that decision makers and/or the general population are not aware of these important differences in outcomes or do not value the economic consequences enough to base their regime change decisions on the possible economic advantages (2011, 8).

[13] On the issue of stability it is important to mention that despite the hopes for rapid and peaceful transitions toward democracy of the political scientists at the beginning of the 1990s, the current stability in many cases was achieved after the entrenchment and the “freezing” of many countries in a gray area of semi-authoritarian, hybrid regimes.

[14] These three cases are considered to represent stable changes from one regime type to another because the changes occurred within the accepted legal framework of the each country. While politics in these countries all displayed a certain level of conflict within the elites, this paper does not consider it to amount to the level of instability that brought down previous regimes. The observation of Fish that “politicians from different branches of government do often accuse one another of bearing responsibility for all of their country’s problems. But they may do so without impugning the political system itself. Most crucially, ordinary citizens can and do fault their least favorite political force — without necessarily regarding democracy itself as the source of their troubles” (Fish 2001, 332).

[15] These dimensions are summarized for each case in Table 1 at the end of the paper.

[16] The regime additionally exploited the ethnic nationalism of the war by resurrecting as patriots the leaders of the pro-Nazi Ustasa regime, adopted its flag, currency, and anthem. Additionally it dropped the official use of the Cyrillic script favored by the Serbs, fired government-employed Serbs, and dramatically reduced Serbian parliamentary representation (Ottaway and Maltz 2001, 376).

[17] Another constitutional amendment in March 2001 eliminated the Chamber of Counties altogether, creating a unicameral parliament.

[18] Some Croatian observers however challenge the view that the Croatian parliament has gained  power after 2000. Ilisin, for example argues that after the constitutional amendments the legislative simply changed masters. It traded a subordinate position for the President of the Republic to a similar spot for the Government, continuing to be a marginal actor in politics. In his opinion, the Croatian Parliament was burdened by an exaggerated symbolic significance, being perceived as “the primary institutional bearer of independent political life” and “a part of the Croatian state and legislative tradition” (2007, 49).

[19] Lucan Way uses the concept of “pluralism by default” to explain the pervasive phenomenon in the former Soviet space where political competition survives “not because leaders are especially democratic or because societal actors are particularly strong, but because the government is too fragmented and the state too weak to impose authoritarian rule in a democratic international context” (2002, 232). Leaders lack the authority to prevent defection of their allies to the opposition, control the legislature, or use force against political opponents. While this ensures continued political pluralism and prevents the concentration of resources in the hands of incumbents, it also “undermines effective governance and may ultimately threaten long-term democratic consolidation” (127). In Moldova this type of pluralism emerged from the polarization over national identity as well as weak elite networks (Way 2003, 455).

[20] As he was also chair of the Constitutional Commission, this made him the designer of the constitutional text.

[21] In explaining the Communists’ electoral success, several important factors come into play: the disastrous socioeconomic situation making Moldova the poorest country in Europe, with 60–70% of its population living below the poverty line; the departure of between 600,000 and 800,000 Moldovans to work abroad in Russia or Western Europe (some 15% of the total population), leaving home an electorate comprised mostly of pensioners, who tend to be the poorest, most nostalgic, most electorally disciplined, and most pro-Communist of all voters. Moreover, the fact that this party not participated in any national elections prior to 1998 meant that they could present themselves as a group of “outsiders,” a tactic that was used actively during a 2001 campaign in which politicians’ integrity was the dominant theme (March 2004, 511).

[22] At the time Mongolian politicians were identifying their political system as a parliamentary democracy though the analysis by Fish (2001, 331) and Munkh-Erdene (2010, 315) shows that the 1992 Constitution was semi-presidential but with a slightly stronger legislature than in classical examples of this regime type.

[23] The MPRP, originally the Communist party founded in 1924, has been in power roughly from its creation to the present except for a brief period between 1996 and 2000. Even after the collapse of Communism and the development of a multiparty system, it quickly transformed itself into a dominant force in Mongolian politics (Rossabi 2009, 231).

[24] Munkh-Erdene (2010) depicts in great detail a particularly vivid portrait of the 1998-2001 political struggles pointing out to the ways in which the interest of the individual political actors shaped the process of regime change.

[25] President Bagabandi initially vetoed these amendments as hasty, ill planned, and as a violation of the Constitution owing to the failure to consult with the Constitutional Court. However, he changed his opinion after being re-elected for a second term in May 2001.

[26] The AKP holds 326 out of 550 parliamentary seats, and, therefore, lacks the supermajority required to amend the constitution unilaterally. The AKP must win the support of at least one more party to rewrite Turkey’s 1982 constitution, which was written under military rule. Recently the AKP came closer to finding an ally on presidentialism through its tentative partnership with the pro-Kurdish Peace and Democracy Party (BDP), which indicated it might accept the AKP proposal in exchange for more comprehensive protections for Kurdish rights (Krueger and Chugh 2013).

Viewed 1974