Constitutional Reforms and Political Turnover in Uruguay: Winning a battle, losing the war

Tags: traditional parties, constitutional reform, Frente Amplio, traditional party, parties, Colorado, Blanco Party, Uruguay, presidential candidates, primary, political parties, FA, candidates, candidate, presidential candidate, Juan Pablo Luna Instituto de Ciencia, DANIEL BUQUET, WINNING A BATTLE, LOSING THE WAR DOCUMENTO, Daniel Buquet Instituto de Ciencia Pol, ICA DE LA FACULTAD DE CIENCIAS SOCIALES DE LA UNIVERSIDAD DE LA REP, MONTEVIDEO, URUGUAY., the 20th century, Daniel Chasquetti, Latin America, American Journal of Political Science, Colorado party, electoral support, INSTITUTO DE CIENCIA POL, References Altman, DAVID ALTMAN, electoral alliance, internal competition, party, electoral victory, presidential office, presidential elections, electoral system, Colorado Jorge Batlle, CONSTITUTIONAL REFORMS, Pontificia Universidad Cat�ica de Chile, electoral processes, internal conflict, DOCUMENTO DE TRABAJO
Constitutional Reforms and Political Turnover in Uruguay: Winning a battle, losing the war David Altman Instituto de Ciencia Polнtica, Pontificia Universidad Catуlica de Chile, [email protected] Daniel Buquet Instituto de Ciencia Polнtica, Universidad de la Repъblica, [email protected] Juan Pablo Luna Instituto de Ciencia Polнtica, Pontificia Universidad Catуlica de Chile, [email protected] 1
і6HUHPRVXna pobre y oscura republiquita, SHURWHQGUHPRVOH\HFLWDVDGHODQWDGLWDVґ Josй Batlle y Ordусez in El Dнa Introduction &RQVWLWXWLRQDO UHIRUPV DUH D FUXFLDO LQWHUYHQLQJ YDULDEOH PHGLDWLQJ EHWZHHQ іKLVWRULFDO FDXVHVґ DW WKH WLPH RI WKHLU LPSOHPHQWDWLRQ DQG WKH TXDOLW\ RI GHPRFUDF\ LQ WKHLU DIWHUPDWK. Consequently, constitutional reforms are endogenous to the will of significant stakeholders at the moment of approval depending on the political-institutional and international context in which they transpire. Simultaneously, as an independent variable and across time, reforms vary in terms of the effects and externalities they produce and the ratchet effects (the irreversibility of enacted reform measures) they create (Pierson 2004). 8UXJXD\ SURYLGHV D ULFK PLOLHX WR H[DPLQH WKH SURMHFW¶V WKHRUHWLFDO IUDPHZRUN. By almost any criteria, the country has been an institutionalized liberal democracy for a significant part of the 20th century, with political conflict and change following institutionalized and democratic procedures. Therefore, recent constitution-making has not been advocated or sought as a vehicle for LQWURGXFLQJEURDGіGHPRFUDWL]LQJґUHIRUPVDVLQ other Latin American cases, particularly in the Andes). However, in spite of being the Latin American country with more years of democratic experience since the turn of the 20th century, no constitutional regime in Uruguay has survived more than 18 years without suffering significant changes. In particular, the country has seen continuous turnover regarding electoral rules and the structure of the executive branch. During the 20th century, these reforms have been usually negotiated and put forth by coalitions of different political fractions benefiting from a contingent convergence of their A previous version of this paper was delivered at the 2006 Annual Meeting of the American Political Science Association, August 31 - September 3; Philadelphia, PA. This research was financed by )21'(&<7¶V3UR\HFW #1060749. 2
(electoral) short-term interests.1 2YHUDOO FRQVWLWXWLRQDO іHQJLQHHULQJґ ZDV IUHTXHQWO\ introduced to isolate the Uruguayan party-system (bipartidist, fractionalized,2 and with a long tradition of co-participation of both parties in government) from socioeconomic change, enabling continuity and stability through electoral and constitutional reforms.3 Against this backdrop, the constitutional reform of 1996 is consistent with the long-term trajectory of the country. This reform pursued an explicit objective for both traditional parties: to prevent Frente Amplio (the leftist opposition to traditional parties) from arriving into presidential office, at a time when the traditional electoral formulas that ensured bi-partisan continuity had stopped working with the creation and continuous electoral growth of Frente Amplio. In the short run, the main objective of the reform was accomplished, with the Colorado party maintaining the presidency in 1999. Moreover, in the short-run, internal 1 7KHUH LV GLVDJUHHPHQW RQ ZKHWKHU WR XVH WKH FRQFHSW µIUDFWLRQ¶ RU µIDFWLRQ¶ LQ WKH FRQWH[W RI WKH Uruguayan party system. Some authors have called these political units factions (see Coppedge 1994: 199; Mainwaring and Shugart 1997b: 425). Following Sartori (1976), we will not use the term faction because it has GHURJDWRU\ FRQQRWDWLRQV LW LV GHHPHG µD SROLWLFDO JURXS EHQW RQ D GLVUXSWLYH DQG KDUPIXO facere¶. Also, we consider this definition misleading since fractions are more permanent than factions (such as the circumstantial µLQV¶DQGµRXWV¶JURXSVIRUPHGLQ9HQH]XHODQSDUWLHV&RSSHGJH0RUHRYHUDQLPSRUWDQWSDUWRISROLWLFDO identities is directed toward fraction rather than political parties. (Altman 2000: ft7 at p.278-279). Even though it is relatively common for political parties to be internally divided into fractions and sometime factions, the peculiarity of the Uruguayan case is that these party fractions show great political visibility, to the point that in many cases they have been considered as real parties inside parties (Lindahl 1977). Nevertheless, Uruguayan parties are still parties, as many scholars have argued (Gonzбlez 1993; Mieres 1992)DQGWKH\DUHWKXVEHVWGHVFULEHGDVіIUDFWLRQDOL]HG SDUWLHVґ (Buquet, Chasquetti, and Moraes 1998). 2 In this paper, party fragmentation refers to the number of political parties in the party system, and party fractionalization refers to the number of fractions within parties (or lemas to be more precise). 3 Until the beginning of the twentieth century, the relations between the two parties were mostly belligerent. Although Uruguayan politics during the nineteenth century were basically centered on fraud, abstention, and uprisings, agreements between political elites were frequent. The so-FDOOHGіWUDGLWLRQDOґSDUWLHV in Uruguay (Colorado and Blanco) are older than the country itself. There is scholarly consensus that these two political groups emerged in the battle of Carpinterнa (1836), when they fought against each other in a civil war, the Guerra Grande, six years after the adoption of the first constitution and eight after WKH FRXQWU\¶V independence. і&HUWDLQO\LQWKHSDVWFHQWXU\WKH\ZHUHQRWSROLWLFDOSDUWLHVLQWKHSUHVHQWVHQVHRIWKHWHUPEXW they were strong political organizations with mass following ±and even armies± and they survived to the SUHVHQWґ(Gonzбlez 1995: 140). 3
debate on the constitutional project triggered significant conflicts within FA leading for LQVWDQFHWRWKHUHVLJQDWLRQRI/LEHU6HUHJQLWKHSDUW\¶s historical leader, in February 1996. 7KHUHIRUH іHOHFWRUDO HQJLQHHULQJґ VHHPHG WR VHUYH RQFH DJDLQ WKH SXUSRVH RI SURWHFWLQJ WKH traditional party-system from its challengers. Notwithstanding, the 1996 constitutional reform process was punctuated by intense negotiations among the traditional parties, the Nuevo Espacio, and some moderate leaders of Frente Amplio. Although such a broad coalition was unnecessary to approve the Constitutional reform in Congress, the project should also be ratified in a popular plebiscite. Therefore, the project should not only be functional to the interest of both traditional parties, but should also seek to accommodate broader interest in order to generate a critical mass of partisan support for the reform. Moreover, facing the challenge of making the reform legitimate and worthy in the eyes of the citizenry, interest should be dressed with reasons. Therefore, while the reform left some historical features of the system untouched (i.e. strict proportional representation, the number of seats in both chambers, the size of electoral districts, and compulsory voting), a series of additional measures besides the introduction of a runoff electoral system to substitute the plurality vote for the presidential election were included: a) the double simultaneous vote (DSV) was eliminated for the presidency and mandatory party primaries were introduced, compelling parties to present only one presidential candidate; b) for congressional elections, deputies were forbidden to build adhoc intra-SDUW\ HOHFWRUDO DOOLDQFH µHOHFWRUDO FRRSHUDWLYHV¶, in Uruguayan jargon) among diverse groups in order to win within the party (i.e. it no longer made it possible for deputies from different fractions within the same party to ally seeking to accumulate their votes); and c) the absolute concurrency of elections (presidential, congressional, and municipal) ended. Whereas the first two of these reforms addressed historical concerns of the left (Frente Amplio and the Nuevo Espacio), the third was sought to benefit the Blanco party by splitting municipal elections (in which this party was stronger than the Colorado) from presidential and congressional races. Is has to be said that, since its maturity in 1942, the Uruguayan electoral system presented a series of characteristics that, taken together, make this a very unusual case in the democratic world (Altman 2002; Buquet, Chasquetti, and Moraes 1998). One of its most original characteristics was the use of the double and triple simultaneous vote, with voters simultaneously electing at two levels: within and between parties (Altman and Chasquetti 2005). All elections in 4
Uruguay used to be connected (vinculadas) and concurrent, making the voter cast a ballot for the same party at all office-levels (see Figure 5 of a ballot paper in Uruguay). Actually, voters simultaneously selected candidates to six representative posts: 1) President and Vice-president, 2) Senate, 3) Deputies, 4) Departmental Major, 5) Departmental Legislators (ediles), and 6) Departmental Electoral Courts. In Uruguay Congress is still elected with proportional representation in both chambers. For the thirty members of the senate the whole country forms a unique national district. Also, in lower-chamber elections (99 members) the entire country is taken as a unique national district for distributing legislators among the lemas. However, Uruguay has 19 districts for distributing deputies among the sublemas and fractions within each sublema. Due to the fact that the 19 circumscriptions vary considerably in their size (from 2 to 45), there is another electoral operation that fits (corrects) the national vote for that list and the number of deputies that each department has.4 In this mathematical correction, it is possible to observe that some departments yield deputies to others in order to maintain the most proportional system of representation possible. It is interesting to note that although Uruguay has two chambers, both overlap their representativeness and both are elected through an integral proportional representation system. The Uruguayan electoral system was designed to maintain a two-party system, allowing fractions withLQSDUWLHVWRFRPSHWHZLWKHDFKRWKHUZLWKRXWKXUWLQJWKHSDUW\¶VFKDQFHVWRJHW elected. In other words, the electoral system was both a cause and a consequence of 8UXJXD\¶V IUDFWLRQDOL]HG ELSDUWLVP. The DSV created a strong incentive for cooperation among sub-lemas, while nationwide PR actuates as a strong incentive for competition among them. Nonetheless, the system did not succeed in avoiding the emergence of a third force. The bipartisan configuration ended in 1971, with the creation of the Frente Amplio (FA) (Gonzбlez 1993: 43). Table 1 below illustrates the evolution of the main characteristics of 8UXJXD\DQFRQVWLWXWLRQVDQGWKHFRXQWU\¶VSUHVLGHQWV 4 This is callHGWKHіWHUFHUHVFUXWLQLRґ 5
Matury of Uruguayan Electoral System
Table 1: Uruguayan presidents and Constitutions since Polyarchy
1919 Baltasar Brum PC 1923 Josй Serrato PC 1927 J.Campisteguy PC 1931 Gabriel Terra PC 1933-1934- Gabriel Terra 1934 Gabriel Terra 1938 Alfredo Baldomir
1919 NCA: Feliciano Viera
1921 NCA: Josй Batlle y Ordусez PC
1923 NCA: Julio M. Sosa
1925 NCA: Luis A. de Herrera PN
1927 NCA: Luis C. Caviglia
1929 NCA: Juan P. Fabini
1931 NCA: Baltasar Brum
1933 NCA: Antonio Rubio
1943 Juan Josй de Amйzaga
1947 Tomбs Berreta
1947 Luis Batlle Berres
1951 Andrйs Martнnez Trueba
1952 NCG: Andrйs Martнnez Trueba
1955 NCG: Luis Batlle Berres
1956 NCG: Alberto F. Zubirнa
1957 NCG: Arturo Lezama
1958 NCG: Carlos L. Fischer
1959 NCG: Martнn R. Echegoyen
1960 NCG: Benito Nardone
1961 NCG: Eduardo V. Haedo
1962 NCG: Faustino Harrison
1963 NCG: Daniel Fernandez Crespo
1964 NCG: Luis Giannattasio
1965 NCG: Washington Beltrбn
1966 NCG: Alberto Heber
1967 Oscar D. Gestido
1967 Jorge Pacheco Areco
1971 Juan Marнa Bordaberry
1976 Alberto Demicheli
1976 Aparicio Mendez
1980 Lieutenant Gral. Gregorio Бlvarez
Interim: Rafael Addiego Bruno
1985 Julio Marнa Sanguinetti
1990 Luis Alberto Lacalle
1995 Julio Marнa Sanguinetti
2000 Jorge Batlle
2005 Tabarй Vбzquez
Constitution of 1917. Semi-Collegiate (Executive power was divided between two bodies, the President and the National Council of Administration). Chamber: 123 (PR); Senate: 19 (PLU) Renovation of NCA every two years. 7HUUD¶V&RXS Constitution of 1934. Chamber: 99 (PR); Sen. 30 (15 y 15). Obligatory participation of both parties in Ministers council. Constitution of 1942. Chamber: 99 (PR); Senate 30 (PR). Freedom for the president in his power of appointment of ministers Constitution of 1952. і3XUHґ&ROOHJLDWH. Executive Power: Consejo Nacional de Gobierno, (composed by 9 members directly elected by citizens. Six counselors for the most voted list of the most voted lema and three for the lema that follows in number of votes but distributed proportionally among its lists).5 Constitution of 1967: Uni-personal presidency. Incorporation of mechanisms of referendum and popular initiative). Dictatorship (1973-1985) (Constitution of 1967 suspended) Constitution of 1967. Constitution of 1996.
5 More on collegial executives could be found at Altman (2008), Sacchi (1999). 6
The reasons on which citizen support for the reform of 1996 was requested derived from some of the anticipated consequences of the four amendments address above. On the one hand, the runoff would enable the crafting of stronger government coalitions, breaking the frequent congressional stalemate that the previous system created. On the other hand, the introduction of unique presidential candidacies, the restriction of the DVS, and the separation of elections would SURYLGHWKHFLWL]HQDEHWWHUFDSDFLW\WRNQRZіH[DFWO\ZKRJHWVIDYRUHGE\ KHUYRWHґZKLOHHQKDQFLQJ her liberty to select candidates from different parties at different governmental levels. Whereas our evidence will show that the first reason on which reform was advocated was unfunded (coalitional politics was further complicated after the reform was introduced), we will claim that the second series of reasons XVHG WR іVHOOґ WKH UHIRUP WR WKH SXEOLF ZDV more accurately predicted and promised. Precisely, the effects of the additional reforms introduced to accommodate the interests of a broad partisan coalition and to lower the prominence of the runoff system within the reform package, ended up producing more harm than good to traditional parties. Interacting with long-term sociopolitical and economic processes, the negative externalities (from the point of view of reformers) produced by the 1996 Constitutional reform significantly favored the electoral chances of Frente Amplio, which reached presidential office in the election of 2004 on the basis of an absolute majority of the popular vote. (ORTXHQWO\IRUWKHILUVWWLPHLQ8UXJXD\¶V\HDUVDVDQLQGHSHQGHQWVWDWHQHLWKHURI the two traditional parties won the presidency. Although such outcome cannot be explained only as the by-product of the unanticipated consequences of the 1996 Constitutional reform (see i.e. the works in Buquet 2005; see i.e. the works in Lanzaro 2004), we will claim that those externalities contributed to reduce the ability of traditional parties to compete with Frente Amplio. The paper is organized around two sections. We first enter to a more detailed description of political negotiation and the reform process leading to the enactment of the 1996 Constitutional reform, providing evidence on the conjunction of interests and reasons that crystallized in the popularly approved reform package. Because of limitations of space, this paper skips a historical constitutional evolution of Uruguay since 1917, focusing on the 7
most important differences existing among WKHFRXQWU\¶VQXPHURXVFRQVWLWXWLRQV 1942, 1952, 1967, and 1996), but this could be found elsewhere.6 In any case, an historical RYHUYLHZ ZRXOG EH LOOXVWUDWLYH RQ KRZ WKH WUDGLWLRQDO XVH RI іHOHFWRUDO HQJLQHHULQJґ LQ WKH country serve as a particular instance of constitution-making as normal politics in Latin America. In this case, those processes of constitution-making were essentially targeted at maintaining the primacy of the traditional party-system and were usually introduced in the context of democratic procedures. The second section explores the long-term consequences of the reform. Finally, we conclude stressing the theoretical contributions of this case to the comparative project on constitutional reforms. 2. Political and Institutional Causes of the 1996 Reforms In this section, we first describe the political context in which the 1996 Constitutional reform was proposed. We then impute interests to relevant political actors and describe the reasons they provided for favoring or opposing the reform. Finally, a description of the enacted reforms is provided. Both interests and reasons are treated as the expected outcomes of the reform. In the third section of the paper, we then explore short and long-term outcomes linked to the reform, assessing the degree to which those expected outcomes were fulfilled. /DFDOOH¶V JRYHUQPHQW - VWDUWHG ZLWK D FDOO IRU D і1DWLRQDO &RLQFLGHQFHґ which once again proved short-lived (Altman 2000). Growing inflation and declining real wages led to a sharp-GHFOLQHRIWKH3UHVLGHQW¶VSRSXODULW\UDWLQJVLQVL[PRQWKVDFKLHYLQJWKH same levels than those shown by Sanguinetti at the end of the previous term. During his first years in office, Lacalle promoted liberalizing and privatizing measures, crystallized in the Ley de Empresas Pъblicas approved in Congress. However, this law was challenged by a referendum promoted by state-unions and Frente Amplio and then supported by Foro Batllista (the Colorado fraction led by Sanguinetti). In December of 1992, the citizenry rejected the law leaving the government without one of its main policy platforms. Additionally, during this period, five proposals to reform the pension system were introduced, with all of them, failing to crystallize in a consensual agreement (Castiglioni 2005). 6 For a more legalistic description of the Constitutional evolution of Uruguay see (Gros Espiell 2002). 8
Meanwhile, resulting from his governing style in Montevideo, Tabarй Vбzquez consolidated as a highly popular leader, becoming a central player for the election of 1994. In those elections, Sanguinetti from the Colorado party succeeded in wining his second presidential election, while Frente Amplio was re-elected to the Municipal Government of Montevideo. However, at the presidential level, this time the electorate was divided in almost equal thirds, with the Colorado Party (PC) obtaining 32.3%, the Blanco Party (PN) 31.2%, and Frente Amplio 30.6%. This electoral outcome catalyzed the creation of a stronger traditional-party coalition after the election. 7KHUHIRUH)UHQWH$PSOLR¶VіDOPRVWYLFWRU\ґSURYRNHGDIXQGDPHQWDOZDWHUVKHGRQWKH FRXQWU\¶VUHFHQWKLVWRU\, increasing the perceived costs of loosing office for both traditional parties. While before 1994 loosing office meant that the other traditional party would win but without a majority, now the implications were far more drastic as both traditional parties would be left out of office. Historically, both traditional parties shared access to the state and were therefore able to distribute patronage and pork on that basis. If neither the Blanco nor the Colorado party were in office, the very sources of both tUDGLWLRQDO SDUWLHV¶ HOHFWRUDO loyalties would be seriously hindered. However, there is a qualitative shift from coparticipation towards coalitional rule. Filgueira and Filgueira (1997) have argued that the incentive structure for the usual іIUHH ULGHU VWUDWHJ\ґ RI WKHVH SDUWLHV KDYH VXEVWDQWLDOO\ FKDQJHG LQ . Such strategy was conducive to frequent decisional stalemates and allowed the traditional party in the opposition to benefit from the votes lost by the incumbent. Moreover, co-participation practices were mostly based on pork distribution not frequently tied, as this time, to an explicit reformist agenda. Interestingly, the coalition between the two traditional parties arose when the growth of the left substantially brought into question their capacity to maintain the presidential office. In this scenario and together with other significant reforms (Social Security and education) both traditional parties promoted changes on electoral rules from a simple plurality election to a runoff electoral system, closing the door on an immediate electoral victory of the left. Thus, in spite of their shrinking electoral support, a second-round electoral alliance between both parties would imply the need for the left to get more than 50% of the vote on its own, in order to gain the presidential office. The rationale behind the establishment of this new instance seems clear, as its most important justification was to prevent the victory of the left. Given that the electoral bases of 9
the traditional parties are ideologically closer each other than to the FA, if forced in a second round to choose between a traditional party against a leftist candidate, they should arguably vote for the former rather than for the latter. Although it implies a counterfactual, it seems extremely plausible to think that, without having changed the rules, the Frente Amplio would have won the 1999 election, as it obtained almost 40% of the votes in the first round (Buquet 2000). In 1999, and for the first time since the transition, the same party was able to maintain the presidency. Although Frente Amplio won the first electoral round, the coalition of Blancos and Colorados was able to defeat the left in the presidential runoff. This time, after successfully contesting the primary against foristas, LWZDV%DWOOH¶VDQG/LVWWXUQ. Although Lacalle successfully won the Blanco primary on the basis of his political apparatus (defeating an adversarial Juan Andrйs Ramнrez who competed on the basis of corruption charges against Lacalle and Alberto Volontй), his candidacy was fatally wounded as a result of that process of internal competition. This led to an historical defeat of the Blanco Party, which only obtained 22% of the vote. Within Frente Amplio, the primary between Vбzquez and Danilo Astori confirmed once again the absolute primacy of the former in the fight for leading the leftist coalition. The Nuevo Espacio (a small center-left party) contested the election on its own, obtaining half of its original support, and quickly allied with Frente Amplio after the first round. The inclusion of this set of additional provisions in the reform, seen as positive steps removing some of the VRXUFHV RI WUDGLWLRQDO SDUWLHV¶ IOH[LELOLW\ DQG DGDSWDWLRQ FDSDcity, moved Astori and Seregni from the Frente Amplio to favor the reform openly contesting 9i]TXH]¶VVWDQFH. This fact fueled important degrees of internal dissent in the party that ended up ZLWK6HUHJQL¶VUHVLJQDWLRQWRWKHSUHVLGHQF\RIthe Frente Amplio and the strengthening of 9i]TXH]¶VOHDGHUVKLS Negotiating the Constitution of 1996 The proposal, discussion, and establishment of constitutional reforms is not a symptom of any particular stressful situation in Uruguay, since the expedient of constitutional reform has been, as previously seen, used many times during the last century as a mechanism to overcome different political stalemates that challenged the traditional party politics. In 10
Uruguay, politics have a strong emphasis on a juridical dimension and its most important political institutions are included in the constitution, with relatively highly detailed regulations. In other words, the constitution must be changed whenever there is a move to introduce some institutional experiment. Even though the political reform was a permanent discussion topic from the very beginning of the democratic restoration (1985), the actual UHIRUPSURFHVVWRRNSODFHGXULQJ6DQJXLQHWWL¶VVHFRQGDGPLQLVWUation (1995-1999). Nonetheless, the requisites to pass the reform were extremely high, requiring the support of two thirds of the legislature and a popular ratification in a plebiscite. So, the amendment process required crafting a great consensus among the political elite. Moreover, after such consensus was built, the proposal should also look legitimate and worth-pursuing for the citizenry. Therefore, reforms needed to accommodate the interests of a diverse set of political actors (parties and fractions), while remaining attractive to the public. The sixth constitutional reform in the history of the country was finally approved in December 1996, with the support of 50.4% of the valid vote, one of the closest results in the FRXQWU\¶V electoral history. That day, citizens ratified the constitutional reforms agreed by both traditional parties and the Nuevo Espacio. Meanwhile, the reform was opposed by the Frente Amplio. The reform introduced significant changes in the rules of the political game; especially in the structure of the electoral system. Additionally, a regulatory framework for partisan activities was included, along with some changes regarding executive-legislative relations. The new electoral rules, first applied for the 1999 electoral cycle, modified three of the four main characteristics of the Uruguayan electoral system. The only feature that was left intact was the use of proportional representation for legislative elections (Espнndola 2001; Finch 1995). First, the plurality system for the election of the President was replaced by the twoballot majority system. Second, the use of multiple simultaneous vote was drastically reduced as now each party could only run one presidential candidate and a maximum of three candidates for municipal mayor. Moreover, the accumulation by sublemas for the election of representatives was eliminated. Third, the all-concurrent election system changed to a partially nonconcurrent system, in which elections were disconnected and separated (the primaries from the general election, the national from the municipal election, and, to a certain extent, the parliamentary from the presidential election). The 1996 Constitution broke down the electoral process into four stages: mandatory party primaries, legislative and presidential elections, a 11
second round or ballotage if no presidential candidate obtained at least 50% of the vote, and municipal elections. A fourth important change was the elimination of the distinction between іSHUPDQHQWґ DQG іDFFLGHQWDOґ lemas, allowing new parties to present a series of lists for the parliamentary election.7 In fact, the electoral system underwent so significant changes that in the long term, massive transformations in the dynamics of the political system can be expected. Additionally, relevant changes were also introduced in regulations affecting the functioning of political parties. First, the restriction forcing parties to present only one candidate to the presidency has brought a fundamental change in the patterns of traditional parties internal competition. Since 1999, all parties should hold simultaneous and mandatory primary and internal elections. The results of the primaries are then subjected to a Sore Loser Law: those who are not elected in this instance cannot run again under other party level in any other election of the electoral cycle. In other words, if someone looses, she must wait for another five years in order to run XQGHUDQRWKHUSDUW\¶VODEHO(Altman and Castiglioni 2006). This rule, also obliges parties to select national conventions of 500 members plus an equal number of substitutes. Additionally, those parties who seek to take part in municipal elections should also elect departmental conventions with a minimum of 50 members. This set of requirements has acted against the survival of tiny parties that used to be part of Uruguayan electoral folklore. Overall, the main strategic goal of the constitutional reform has been achieved insofar as the new electoral rules, specifically the majority run-off, favored the maintenance of the reformist coalition in office. But things turned out differently to what had been originally planned. The winning party was the PC and not the PN, and the majority factions within the parties were radicals rather than moderates. The difference between the actual outcome and what was expected stems from a number of erroneous calculations in the original plan, and some strategic mistakes made during the electoral campaigns (Buquet and Piсeiro 2000). The remainder of this paper is devoted to explain the short and long-term effects of the 1996 Constitutional reform. 7 From 1966 until the 1994 election, only "permanent" parties, i.e. those that already has parliamentary representation, had the prerogative to present more than one candidate/list per party, that is to say to benefit from the multiple simultaneous vote. 12
3. Causes and Effects of the 1996 Reform In this section we contrast the reasons and interests of reformers with the short and long-term effects of enacted reforms. As we shall see, several externalities (frequently contradicting the reformers intentions) can also be observed in this case. We organize this section around three bundles that combine specific reforms enacted in 1996: a) the introduction of a runoff system, b) the introduction of primaries, unique-presidential candidacies, and partially non-concurrent elections; and c) the restriction of the DoubleSimultaneous Vote. For each bundle we identify the expected effect of the reform (from the point of view of reformers), the observed effect (in the short and long run), and the presence and character of externalities. Table 2 summarizes the argument we present in this section. 3.1 The Majority run-off Interests and Reasons Historically, both traditional parties shared access to the state and were therefore able to distribute patronage and pork on that basis. If neither the Blanco nor the Colorado party were LQ RIILFH WKH YHU\ VRXUFHV RI ERWK WUDGLWLRQDO SDUWLHV¶ HOHFWRUDO OR\DOWLHV ZRXOG EH VHULRXVO\ hindered. In this scenario and together with other significant reforms (social security and education),8 both traditional parties promoted an electoral reform changing the electoral rules from a simple plurality election to a runoff electoral system, closing the way for an immediate electoral victory of the left. Thus, in spite of their shrinking electoral support, a second-round electoral alliance between both parties would imply the need for the left to get more than 50% of the vote on its own, in order to gain the presidential office. 8 For a detailed description of these reforms see (Castiglioni 2005). 13
Table 2: Mapping-up incentives, reasons, and externalities 14
Different reasons were advocated in favor of the run-off. Some were very explicit about WUDGLWLRQDOSDUWLHV¶SULPDU\LQWHUHVW. For instance, the Colorado Senator Luis Pozzolo bluntly stated: і(Q8UXJXD\VHSURGXFtDXQDYRWaciуn entre dos partidos que, aunque tienen filosofнas distintas, representan en lo institucional, como respeto democrбtico, ODV PLVPDV UDtFHV SRUTXH QDFLHURQ MXQWRVґ. [Pero las cosas han cambiado,] іDKRUD SXHGH JDQDU HO JRELHUQR GHO SDtV XQD IXHU]D GLIHUente; el paнs puede llegar a tener, si no hace esta reforma, un gobierno de signo marxista. Y esto HVORTXHWHQHPRVTXHSHQVDUґ%~VTXHGD1S. In a somewhat more veiled way, Ope Pasquet (former deputy of the PC) argued that the runoff would seUYH DV DQ HOHFWRUDO LQVWUXPHQW WR DOORZ %ODQFRV DQG &RORUDGRV WR іYRWH WRJHWKHUґіYRWDUMXQWRVґ(Bъsqueda N° 844, 23/05/1996, p.6). In general, more prominent traditional party leaders argued for the reform on the basis of more sophisticated reasons, usually stressing the capacity of the runoff to induce more effective coalition building and governance. For instance, former President Luis Alberto Lacalle argued that: і>[email protected] su base de apoyo [dedicбndose a] conquistar, enrolar, negociar y conceder ciertas cosas a otras fuerzas polнticas para fabricar de antemano una coaliciуn que le DVHJXUHHOJRELHUQR>«@De manera que en el ballotage lo que se plebiscita es ODFRDOLFLyQDVtDUPDGDґ(Lacalle, Bъsqueda N° 867, 31/10/1996, 13). In this line, promoters of the reform argued that single presidential candidates and less fractionalized parties would induce more disciplined legislative behavior and greater incentives for coalition building. Former President Julio Marнa Sanguinetti argued this point stating that: іTodos los partidos estuvieron de acuerdo en que se debe procesar una reforma y ella es necesaria ante tiempos nuevos, con problemas distintos (y) con una enorme fragmentaciуn electorDO >«@ Es necesario ajustar HVWUXFWXUDVґ%~VTXHGD1S Conversely, most leaders of the left opposed this reform on the basis of one simple institutional reason. As Senator Korzeniak (Socialist) warned: іUn ballotage que traiga a un presidente con 55% o 60% de apoyo y un Parlamento (en el) que tiene 30% de apoyo puede generar una situaciуn SROtWLFDGHHQIUHQWDPLHQWRTXHFRQGX]FDDVROXFLRQHVDXWRULWDULDVґ%~VTXHda N° 867, 31/10/1996, p. 13). Likewise, Tabarй Vбzquez alleged that: 15
і&Ueamos la figura del presidente emperador. Esto es peligroso para cualquiera que sea el presidente de la Repъblica, porque va a tener el 51% de los votos con un Parlamento dividido, va a haber un enfrentamiento de poderes, el Poder Legislativo va a estar disminuido en sus poderes, vamos a tener un super-SUHVLGHQWH \ HVWR HV FUHDU FRQIOLFWRVґ %~VTXHGD 1 7/11/1996, p.10). Moreover, FA leaders essentially campaigned unveiling the short-term interest (and its consequences) hidden behind the reasons advocated by traditional party leaders. Vбzquez foresaw the short-term outcomes of the reform brilliantly: іNo le quepa a nadie dudas que si gana la elecciуn el Encuentro Progresista [Frente Amplio] como seguramente las va a ganar en 1999, y esta es la gran preocupaciуn por esta reforma constitucional que es mбs una reforma electoral, y sale segundo un partido tradicional, van a acumular los partidos tradicionales en la segunda vuelta para la candidatura a la Presidencia. Esto lo han dicho los propios dirigHQWHV GH ORV SDUWLGRV WUDGLFLRQDOHVґ 9i]TXH] Bъsqueda N° 870, 21/11/1996, p. 7). Short-term effects: Winning the 1999 Election but without Improving Coalition Building and Governance (1999-2005) The runoff for the presidency was one of the most relevant of the changes introduced in the 1996 reform, if not the most. As previously explained, the short term interest of reformers was to avoid having the left arriving into the national executive. This objective was fulfilled in the short-run, with Jorge Batlle (coming second in the first round) defeating Tabarй Vбzquez from the Frente Amplio in the runoff. At that point, Batlle received the support from leaders of the Blanco Party. The second advocated reason for reform was to improve coalition building. This reasoning involved the assumption that by inducing an electoral coalition for the second presidential round, the reform will definitely strengthen the presidential coalition. This would also be reinforced by other expected effects of the runoff (and a series of other reforms introduced in 1996) as the reduction of the effective number of parties which would put a halt to partisan fragmentation. We now explore the effects of the reform in regard to these objectives. A first reading of the introduction of the runoff system from the point of view of coalition building would indeed suggest that this new institutional device would strengthen post-electoral coalition formation and party discipline. Yet an overlooked fact suggests 16
exactly the opposite. In order to gain office, a traditional party only had to have a bigger share than the other in the first round. As it was unthinkable that the other traditional party would then shift its vote to the left and given vote share distribution among the three parties, office was secured. Thus, while in one sense coalitions should have become more stable, other forces and logics pushed exactly in the opposite direction. The end result was a coalition in which neither partner wanted to stand out with unpopular measures and especially with fiscal adjustment. Eventually, this coalitional configuration ended-up contributing to deepen the FRXQWU\¶VILQDQFLDOFROODSVHLQZKLFKLQWXUQSURYHGIXQGDPHQWDOLQFDWDO\]LQJSRSXODU support for the opposition in the subsequent election (the Frente Amplio). With regard to the effects of the runoff on the effective number of parties (N) and candidates, extant research consensually claims that N is systematically higher when that institutional device is used in presidential elections.9 Therefore, in theory, Uruguayan politicians were mistaken in expecting less fragmentation to occur under the new system. However, strictly speaking, under the new majority runoff system party fragmentation did not increase in Uruguay. In fact, it decreased from 3.4 to 2.5, contradicting the theory and PDWFKLQJ WUDGLWLRQDO SROLWLFLDQV¶ H[SHFWDWLRQV VHH )LJXUH . Nonetheless, a more careful reading of election results shows that fragmentation did in fact increased (Buquet 2004). In short, the observed fall in the magnitude of N could be better explained as the result of a tendency that begun before the reform was introduced and could not be attributed to a direct effect of the new electoral system. 9 See Shugart and Carey (1992), Jones (1995), and Mainwaring and Shugart (1997a) on this point. 17
Figure 1: Effective Number of Parties
4 C onstitution of 1942
C onstitution of 1952
C onstitution of 1967
C onstitution of 1996
2 1942 1946 1950 1954 1958 1962 1966 1971
1984 1989 1994 1999 2004
Source: DXWKRUV¶HODERUation based on information from the Facultad de Ciencias Sociales, Universidad de la Repъblica, Data Bank.
To illustrate this claim, the impact of the new electoral system can be better observed by comparing the electoral results of the national elections for which the majority run-off system was used with the electoral results obtained in the subsequent local elections, in which the plurality system was maintained. The comparison is reasonable because parties and voters were the same in both contests, within a short time span. The next table shows the values of N for the 19 local electoral districts in Uruguay, both for national and local elections held after the electoral reform. When compared to the results obtained in the corresponding national election, the value of N is strictly less in 37 of the 39 local elections. Therefore, we can see a clear pattern of strategic electoral concentration in local elections that did not occurred at the national level. This could happen because the plurality system poses incentives to vote strategically that the majority run-off system does not promote. Furthermore, we have classified departments into two sub-JURXSV іPRGHUQґ DQG іWUDGLWLRQDOґWKDWVKRZGLIIHUHQWSDWWHUQVRIHYROXWLRQLQWKHUHGXFWLRQRIWKHLUUHVSHFtive N's due to the growing trend of the FA (Table 4 below). %\іPRGHUQGHSDUWPHQWVґZHXQGHUVWDQG those where the left has had historically its best electoral performances. This group shows a clear descending trend in the number of parties due to the electoral growth of the FA. By іWUDGLWLRQDOґ GHSDUWPHQWV ZH XQGHUVWDQG WKRVH ZKHUH WKH WUDGLWLRQDO SDUWLHV VWLOO UHWDLQ DQ important electoral share. These ones show a clear contrast in the number of parties contesting local and national elections. While for national elections and under a majority runoff N is 18
higher, at local elections the number falls down to form a strict bipartisan system. In this case and besides the national trend of reduction in the number of parties, we can observe the strategic incentives introduced by the plurality system. Indeed, the observed reduction of N (from 3 to 2.6 in the effective number of parties) in between national elections (1999 and 2004) occurring in traditional departments can be solely explained by the electoral growth of FA. Meanwhile, the 2.1 N observed for both local elections (2000 and 2005) is obviously caused by strategic voting.
Table 4: Effective Number of Parties by Uruguayan departments
Department Montevideo Canelones Maldonado Rocha Salto Paysandъ Rнo Negro Soriano Colonia Florida Mean "moderns" Treinta y Tres Cerro largo Rivera Artigas San Josй Flores Durazno Lavalleja Tacuarembу Mean"traditionals"
1999 2,7 3,2 3,2 3,2 3,1 3,1 3,1 3,1 3,3 3,2 3,1 3,1 3,1 2,9 2,9 3,2 3,0 3,0 3,0 3,0 3,0
2000 2,3 2,6 3,0 2,7 2,9 2,8 3,0 2,9 2,8 2,9 2,8 2,3 2,0 2,5 2,3 1,8 1,5 2,4 2,4 1,8 2,1
2004 2,2 2,4 2,5 2,6 2,6 2,3 2,6 2,5 2,6 2,7 2,5 2,5 2,5 2,8 2,6 2,5 2,6 2,7 2,7 2,4 2,6
2005 2,2 2,0 2,2 2,3 2,9 2,3 2,3 2,3 2,2 2,6 2,3 2,4 2,1 2,6 2,9 2,0 1,7 2,2 1,8 1,7 2,1
Therefore, we can argue that the decreasing N observed after the implementation of the majority runoff system is not an effect of this institutional devise. Conversely, our evidence shows that if the continuity of the plurality system could have granted a greater decrease in fragmentation than the one obtained after the 1996 reform was implemented. Another explicit objective for introducing the runoff was to avoid Presidents having to govern with minority legislative support. However, the combination of single candidates and the majority runoff system that the electoral reform set up, could contribute to worsening the
parliamentary position of the elected President, and this is in fact what happened the first time it was put into practice. Figure 2: presidential legislative contingents
Presidential legislative contingents
0% 1943 1947 1951 1955 1959 1963 1967 1972 1985 1990 1995 2000 2005
President's party
President's faction
Source: Authors' elaboration based on information from the Facultad de Ciencias Sociales Data Bank.
The old Uruguayan electoral system made for at least two absolute certainties: the party of the President would have a plurality (at least) in parliament, which meant that it would KDYHPRUHSDUW\UHSUHVHQWDWLYHVWKDQDQ\RWKHUSDUW\DQGWKDWWKH3UHVLGHQW¶VIDFWLRn would have a plurality within his own party as well. Congress-members directly supporting the President had, up to that time, necessarily been the biggest group in the biggest party. This was guaranteed by the connection existing between presidential and legislative elections. In short, to win the Presidency it was simultaneously necessary to belong to the party that received the most votes for the legislature and to head the fraction within that party that received the largest plurality. Under the current rules, the legislative representation of the party in government does not have to be the biggest in parliament because there is nothing to stop the election of the candidate of the second party, whose representation would be second in number of legislators. This happened in the 1999 election. Along with the reduction in the legislative weight of the 3UHVLGHQW¶V SDUW\ ZKLFK FDQ UHVXOW IURP WKH PDMRULW\ UXQRII (Jones 1995) there is no disposition in the new constitution which guarDQWHHV OHJLVODWLYH ZHLJKW WR WKH 3UHVLGHQW¶V faction. The President elected is the only candidate from his party, and he can be voted along with any of the parliamentary lists of that party. 7KHYRWHVWKDWWKH3UHVLGHQW¶VIDFWLRQREWDLQV 20
are relatively independent from the electoral potential he has, and consequently, he may be in a minority position within his party. Therefore, nothing ensures that the President will enjoy a majority in her party because there is nothing that could stop fractions that were defeated in the internal party elections from emerging triumphant in the legislative election. Main Externality of the runoff system: Consolidating a bi-polar logic of partisan competition With the resumption of democratic politics the main features of the traditional system were restored. However, facing a growing and powerful leftist opposition and confronting increasing public deficits and inflationary pressures, both traditional parties embarked on reformist agendas. The fiscal crisis of the state also limited the scope for reproducing loyalties on the basis of traditional clientelistic and patronage appeals, which became increasingly inefficient for competing with a leftist party that had programmatically appropriated the defense of batllismo and was able to mobilize discontent against neoliberal reformers. Although the Uruguayan party system lacked a strong advocate for neoliberal reform in spite of the specter of a crumbling economy and increasing international pressure, once in office the traditional parties embarked on gradual attempts at state-reform. Given the omnipresence of batllista or statist/redistributive ideology and significant policy feedback from ISI, these reforms were, not surprisingly, extremely unpopular. In this context, the Frente $PSOLRJDYHSROLWLFDOH[SUHVVLRQWRDіYHWR-FRDOLWLRQґRI,6,EHQHILFLDULHVZKLOHDWWKH same time, drawing on a reinterpretation of batllismo to attract votes from the sectors that became increasingly alienated from the traditional system in the wake of tKH іKDSS\ 8UXJXD\ґ FULVLV. Ideologically, this coalition put forth a statist platform advocating and enacting legislation (through the use of direct democracy mechanisms) to roll-back reformist legislation in favor of the status-quo (strong State intervention in the domestic economy). Additionally, while the two traditional parties faced surmounting popular discontent and managed increasingly constraining budgets in order to maintain their clientelistic political machines, the Frente Amplio was able to gradually become the most popular political party in the country. The strength of the frenteamplista subculture solidified through exile and resistance to the repression and persecution during the bureaucratic-authoritarian regime, combined with the loosening RI %ODQFR DQG &RORUDGR¶V SDUWLVDQ LGHQWLWLHV SURYLGHG IXUWKHU room for the transformation of the party-system. 21
This dynamic, coupled with electoral results and the constitutional reform of 1996, crystallized two ideological families competing for votes based on programmatic appeals along the state-market divide. 7KRVHWZRіLGHRORJLFDOIDPLOLHVґFRQVROLGDWHGDEL-polar logic of competition among three significant parties: the Blancos and the Colorados (in government) and Frente Amplio (in opposition). This competitive dynamic reduced the capacity of both traditional parties to keep their programmatic appeals differentiated. Therefore, while before the traditional party not occupying the presidential office could simultaneously benefit from co-participation while opposing the incumbent once the election approached, the runoff system made coalitional arrangements more explicit (though, as argued above, this did not get translated into greater governance efficiency). This helped to consolidate the formation of two clearly distinct (and more or less internally homogeneous) blocks: the government and the opposition; strengthening the latter at a time when economic discontent was on the rise. 3.2 Primaries, internal party elections, unique presidential candidacies, and split elections Interests and Reasons Confronting the need to approve the reforms through a popular plebiscite, and having failed to do so in 1994, reformers needed to ensure a wide basis of support for this initiative. In this respect, the introduction of party-primaries and single presidential candidacies were advocated as natural counterparts of the majority runoff system. These particular reforms also addressed a long-standing claim of the left.10 ,QGHHG іPRGHUDWHVґ ZLWKLQ )UHQWH $PSOLR supported the reform due to this fact. For instance, Danilo Astori, leader of the largest fraction of Frente Amplio in the elections of 1994 persisted in favor of: іDQLTXLODU ODV WUDPSDV (el DVS y el uso de candidaturas presidenciales mъltiples) que, pulverizбndolo, alejan al sistema polнtico de la gente, es verdaderamente un objetivo revolucionario, en el sentido estricto del FRQFHSWRґ%~VTXHGD1S 10 As a matter of fact, the reforms of 1994 failed in part because the leftist parties accepted the primaries but not the balotage. 22
Indeed, the historical leader of Frente Amplio perfectly acknowledged the short-term effects of the reform: іDebemos de entenderlo como una lуgica de la vida polнtica y de la situaciуn polнtica [...] que nuestros adversarios no quieran que nosotros alcancemos el gobierno y el poder y mantenerse ellos en el ejercicio del mismo. [...] їO es que alguno de nosotros piensa que blancos y colorados van a pavimentar de pйtalos de rosa nuestro camino de acceso al gobierno y al poder? (Liber Seregni, Bъsqueda N° 830, 08/02/1996, pбg. 4). Nonetheless, Seregni also supported the 1996 initiative due to the elimination of multiple presidential candidacies and the restriction of the DSV (we address the latter below). Interestingly, from within traditional parties, these elements of the reform were not consensually supported. For instance, Alem Garcнa (a Blanco congreso-member) argued that: і6e propone un centralismo electoral, aumentar el poder de las cъpulas partidarias y, prбcticamente, la eliminaciуn de los dirigentes medios y GHSDUWDPHQWDOHVґ%~VTXHGD1S Meanwhile, the separation of the municipal elections, which allowed voters to cast votes for different parties at different political levels, also provided greater legitimacy to the reform. In particular, this piece of legislation fulfilled the interest of the PN, which had deeper roots than the PC at the local level, in the interior of the country. Short-term effects: Passing the reform and increasing voter-freedom Reformers needed to create a wide basis of partisan and popular support to ensure that the runoff was enacted. These additional measures collected a great deal of support in the partisan spectrum, as well as in the public eye and provided the reform the necessary legitimacy to be (barely!) approved. Moreover, dissent within Frente Amplio launched an internal crisis, that in the short-term, might have also favored the electoral chance of traditional parties. Moreover, presidential elections with one single candidate per party along with the elimination of DVS for lower-chamber elections, have evidently given the electorate greater certainty with respect to the consequences of its vote. The splitting of the electoral cycle also made for a notable increase in voter freedom since it allowed them to opt for different parties at different governmental levels. Long-term effects: Are Primaries an effective mechanism for selecting presidential candidates? 23
It is hard to argue that mandatory party primaries provide a better mechanism for selecting candidates than the DSV (Buquet 2003). The constitutional reform of 1996 removed the double simultaneous vote for presidential elections, forcing parties to present single presidential candidates. In order to maintain the open competition that characterized traditional parties, the amendment established mandatory open primary elections. Since the new rules are in force, all parties ZDQWLQJ WR SDUWLFLSDWH LQ WKH HOHFWRUDO SURFHVV PXVW KDYH іLQWHUQDOґ HOHFWLRQV RQ WKH VDPH date.11 Internal elections serve two purposes: i) selecting the presidential candidate at each party, and ii) electing delegates to the national convention and the 19 departmental party conventions. The National Party conventions are comprised of 500 members (elected by proportional representation) and have two purposes: a) to select the presidential candidate if no candidate received more than 50% of the popular vote, or more than 40% with at least a 10 point margin over the second front runner; and b) to select the vice-presidential candidate. At first sight, the electoral reform would not seem to have generated any significant alterations from the previous model relating to the mechanisms of presidential candidate selection, since as before the fractions could designate pre-candidates. There are numerous forces at work, however, that will change the competition within and among parties. First, the rules to select the party presidential candidates have changed; under the new rules, candidates must win a primary by gaining either 50% of the vote or at least 40% with an advantage of 10 points over the second front runner. If no candidate meets these requirements, a party convention chooses the candidate. This high threshold for winning the primary with the possibility of having the candidate chosen through the convention could yield a more fragmented system, as those observed under two-round electoral systems (Mainwaring and Shugart 1997b). In short, primary elections force candidates belonging to the same party to compete among themselves, highlighting their differences and potentially generating internal confrontations. Here WKH SULVRQHU¶V GLOHPPD VWUXFWXUH RI LQFHQWLYHV LV FOHDU DV FRRSHUDWLYH VROXWLRQV DUH difficult to find. The way a party handles the primary competition, then, has a great impact on 11 Citizen may vote in any party primary, but they may only participate in one. Part of the reasoning behind holding the primaries on the same day for all parties was to prevent voters from intervening in the affairs of more than one party. 24
the subsequent general election. It is normally accepted that parties that contain internal conflict do better in general elections. 7KLVLVQRWGLIIHUHQWIURPWKH86V\VWHPZKHUHіDOO studies point toward divisive nomination campaigns contributing a negative effect to general HOHFWLRQRXWFRPHVґ(Atkeson 1998: 257). Kenney and Rice (1987: 31) DOVRIRXQGWKDWіZKHQ RQH SDUW\ KDV D GLYLVLYH SULPDU\ VHDVRQ ZKLOH WKH RWKHU SDUW\¶V QRPLQHH LV HVVHQWLDOO\ uncontested, the dividHGSDUW\ZLOOEHDGYHUVHO\DIIHFWHGLQ1RYHPEHUґ8UXJXD\DQHYLGHQFH does not yet confirm these findings, but rather seems to suggest the existence of something like an optimal level of competitiveness in primaries, beyond which either higher or lower competition could damage the party. The three main parties have chosen different strategies to deal with the new primary system, and each party dealt with the situation differently in 2004 than they had in 1999. Most notably, there was a significant decrease in the primary competition in the more recent election. The FA had a somehow symbolic internal competition in 1999, but put forth only a single candidate in 2004. The PC had had a well-balanced competition between two candidates in 1999, but also put forth a hegemonic candidate in 2004. The PN, which in 1999 had been excessively fragmented, ran just two main candidates in 2004 who competed without high levels of vitriol. If we link the degree of competitiveness of each primary contest to the electoral performance of each traditional party at the general election, we can see that when the effective number of candidates was close to 2 the party did well, and when this number moved away from 2, either up or down, the performance in the general election worsened (Table 3). This does not seem to be true for the FA, which had a very low level of competitiveness in both elections while their electoral growth continued an upward trend that started in 1971.
Table 3: Effective number of presidential candidates in the traditional- parties primaries
Election Year 1999 2004
Partido Colorado
Ef. N° Candidates* Electoral Growth**
Partido Nacional
Ef. N° Candidates* Electoral Growth**
* Effective number of presidential candidates at the primary contest
** Percentage of electoral growth related to the previous election
In sum, the notion of a well-balanced primary contest implies that in addition to running an effective number of candidates close to 2, the parties should work to minimize confrontation between the two front-runners. Therefore, along with an attractive primary contest, the party must offer a united image to avoid losing voters as the result of a conflictridden primary. While too much confrontation can lead to an internal deterioration of the party, manipulating primaries to avoid confrontation has also been harmful for traditional parties. In 2004 the major fractions of the PC unified behind a single candidate, thus distorting the primary process. The result, however, proved disastrous as the party lost badly in the general election. Again, since the DSV forces intra-party rivals to focus their campaigns more on inter-partisan rivalries, it allows internal party choice without yielding strenuous and divisive internal fights. The DSV, then, favored these parties, whereas primaries may damage them. The DSV presented an additional benefit both for parties and voters. Josep Colomer VKRZVKRZіWKHFDQGLGDWHVVHOHFWHGRQDSULPDU\HOHFWLRQV¶EDVLVWHQGWREHQRWYHU\SRSular RU ORVHUV LQ WKH FRUUHVSRQGLQJ SUHVLGHQWLDO HOHFWLRQVґ (2002: 119). This occurs due to the differences in who participates in the two rounds. 5HVRUWLQJ WR D VSDWLDO PRGHO &RORPHU¶V reasoning can be clearly illustrated. The winning pre-candidate of a primary election should be the one who is the closest to the median voter within the partisan electorate. However, a centrist within a party may be an extremist for the whole electorate. A leftist party, therefore, would do better in the general election by choosing a candidate who is to the right of their median member and a rightist party would be best positioned by choosing a candidate to the left of their overall membership. The DSV resolves this problem for the parties too, in that it allows the parties to present a wide array of options to the whole electorate, therefore benefiting both the citizenry, which enjoys a wider menu, and the party, which does not limit its electoral appeal. $OWKRXJK WKHUH DUH RQO\ WZR HOHFWLRQV WR FRQVLGHU 8UXJXD\¶V UHVXOWV RQO\ SDUWLDOO\ support these hypotheses. ,Q WKH 31¶V ZLQQLQg pre-candidate ±Lacalle± was without doubt closer to the median Blanco voter, but his defeated contender ±Ramнrez± was much FORVHUWRWKHZKROHHOHFWRUDWH¶VFHQWHU. Something similar could be said about the FA, whose winner, Vбzquez, clearly showed a more radical profile than the defeated Astori. The winner in the Colorado internal election, Batlle, however, was ideologically placed more to the right than the defeated candidate, Hierro, but Batlle was still in a much better position with the entire electorDWH GXH WR +LHUUR¶V UHODWLRQ ZLWK WKH RXWJRLQJ JRYHUQPHQW. In 1999, then, the 26
Colorados chose the best candidate for the general election and were successful. In 2004, the only party with a true primary was the PN, choosing Larraсaga over Lacalle. In this case Larraсaga was closer than Lacalle to the median voter; thus, contradicting &RORPHU¶V002) claim that party militants would choose more extremist candidates. Even though the Blancos chose a centrist, the Frente Amplio won the general election in a landslide, in spite of running Vбzquez who is positioned on the left of the ideological scale as the presidential candidate. In sum, the primaries have not always yielded winners who are more representative of their parties than the full electorate, and it is also unclear that these relatively extreme candidates are poor general election competitors, perhaps in part because the other parties also choose poorly. Moreover, holding primaries independently of the general election produced to additional consequences: i) it lengthens the political campaigns and therefore ii) it increases the costs of campaigns and electoral processes. It is generally assumed that these are undesirable consequences for a political system, especially in a country where political or government spending is of high concern. The high costs of the electoral organization and the prolonged nature of campaigns can cause loathing and political distrust in the citizenry, thus KDYLQJDQDGYHUVHHIIHFWRQWKHV\VWHP¶VOHJLWLPDF\. At least one important advantage of the DSV system was its shorter campaign season due to the simultaneous holding of primary and general elections. The legislature did try to deal with the long campaign period by moving the primaries back from April to June, by a law in 2004. Primaries in April forced the parties to define their pre-candidates the previous year, and the move to June has resulted in a shortening of the calendar by considerably more than just two months. Main externality of the new election system: The increasing detachment between local and national partisan structures: A trend towards the de-nationalization of elections was observed in Uruguay, particularly in 1999-2000, with local electoral contests presenting diverging trends from national elections in some districts. Specifically, whereas the Blanco and Colorado parties benefited from such trend, the FA lost some of the electoral support it received in the presidential election (Guerrini 2000; Magri 2000). This can be explained by the separation of national and local elections, the consolidation of strong local leaderships, and the progressive popular disenchantment with national ones. In this context, Mayors gained autonomy and 27
ZHUHEHWWHUDEOHWRGHFLGHZKHWKHURUQRWLWZDVFRQYHQLHQWIRUWKHPWRіSXWWKHLUDSSDUDWXVHV DWSOD\ґIRUWKHQDWLRQDOHOHFWLRQ. This trend was strengthened by the constitutional provision that established unique presidential candidates for every party contesting the election. In cases where supporting a national presidential candidacy seemed openly LQFRQYHQLHQW WR D ORFDO OHDGHU JLYHQ WKH SUHVLGHQWLDO FDQGLGDWH¶V ORZ OLNHOLKRRG RI ZLQQLQJ RIILFH DQG WKH 0D\RU¶V DOLJQPHQW ZLWK D GLIIHUHQW IUDFWLRQ IURP WKH RQH WKDW KDG ZRQ WKH primary (and presidential nomination) of their party (e.g. the case of Larraсaga in Paysandъ, Da Rosa in Tacuarembу, or Cerdeсa in San Josй), Mayors did not decisively help national leaders in their campaigns. After the first round, covert negotiations between local and national activists of both traditional parties also took place. For instance, in San Josй, Juan Ciruchн a former and extremely popular Herrerista mayor with a great personal following in the district exchanged his support for the candidacy of Colorado Jorge Batlle in the presidential runoff for Lista 15 support in his electoral battle against his former Municipal Secretary and incumbent Mayor: Cerdeсa.12 In 2005 however, likely following the national wave favoring FA, the levels of nationallocal divergence were lower, granting the left eight Mayoralties (five previously held by the Blanco Party and three corresponding to the Colorado one), with first time electoral victories in municipal contests outside Montevideo. In turn, while the Colorado Party was only able to win (maintain) one Mayoralty, the Blanco Party obtained ten. 3.3 The elimination of electoral cooperatives Interests, Reasons, and Short-Term Effects Eliminating the accumulation by sublemas was in the interest of the main national fractions, which could, as Garcнa anticipated, recentralize power within parties and eventually induce greater levels of congressional discipline and more effective governance. Nonetheless, the main short-term interest of reformers was the same than the one pursued through the reforms described in b; namely: to widen the potential political and electoral support for the LQLWLDWLYH &LWL]HQ¶V SRRU RSLQLRQ RQ іHOHFWRUDO FRRSHUDWLYHVґ DQG )UHQWH $PSOLR¶V KLVWRULFDO 12 Interviews with Juan Chiruchн, Miguel Zunino, and Jorge Cerdeсa in San Josй (2003). 28
request regarding their elimination supported this rationale. Indeed, while defending the reform proposal, Danilo Astori from the Frente Amplio also attacked the effects of the triple DVS: і>«al sumar sus votos para perjudicar a otra mayorнa] las cooperativas electorales WHUJLYHUVDQ PXFKR OD YROXQWDG GHO HOHFWRUґ %~VTXHGD 1 28/11/1996, p. 72). The approval of the Constitutional reform in 1996 fulfilled the main short-term interest pursued by reformers. In the long-run, the additional interests of reformers were also fulfilled, leading to the centralization of power by national factions and a more direct linkage between upper and lower chamber lists.
Long-Term Effects: Greater Transparency, a Decreasing Number of Lists for the House, and Greater Centralization of Power within Traditional Parties. The elimination of the accumulation of sublemas in the election of representatives has made for a very much simpler choice for the electorate, and this can be seen in the reduction in the number of lists presented in 1999 and 2004, which amounted to less than a half of those presented in 1994 (see Figure 2). As long as different lists for the House cannot accumulate their votes, the smaller ones have either to join one of the biggest or to build a single list among several of them.
Figure 3: Number of Ballot Lists for the Chamber of Deputies
Number of ballot lists
0 1942
Source: DXWKRUV¶HODERUDWLRQ based on information from the Facultad de Ciencias Sociales Data Bank.
Moreover, the reduction in the legislative supply is associated with the predominance of big national fractions over local political groups. Under the new system, a rigid connection between the supply for the Senate and the supply for the House of Representatives is enforced. Consequently, we can expect more disciplined legislative behavior on the part of lower-chamber representatives.
Figure 4: Effective Number of Senate Lists
Effective number of senate lists
0 1942
Partido Colorado
Partido Nacional
Frente Amplio
In spite of the reduction in the number of lists of candidates, which has enormously reduced and simplified the supply to the electorate, the internal fractionalism of the political parties has not gone the same way. In this area, the reform does not seem to have had any particular impact, since the internal fractionalism of the Frente Amplio increased, that of the Blanco Party has been reduced, and that of the Colorado Party has held steady. As long as electoral rules for the Senate did not change we should not expect any changes in party fractionalism, but it could be argued that single presidential candidates pose incentives to more fragmented parties as long as the competition for president and for senate seats was disconnected (Buquet 2001). 7KHJURZLQJWUHQGLQ)$¶VLQWHUQDOIUDJPHQWDWLRQFRXOGUHIOHFW such a process, while the opposite process at the traditional parties could be better explained due to their declining share of seats.
Main externality of the elimination of electoral cooperatives: The weakening of traditional SDUWLHV¶WHUULWRULDODSSDUDWXVHV The simultaneous election of national and departmental party authorities in internal elections, combined with the new timing of elections and the elimination of the proliferation of lists in parliamentary elections have altered the internal balance between national and local partisan/fractional organizations. The joint effect of electoral reforms and the economic crisis that started in 1998 and got deepened during 2002 and 2003, have weakened the two most prominent national political apparatuses of the traditional parties (the Herrerismo in the Blanco Party and the Foro Batllista in the Colorado Party). It seems possible to claim that traditional parties have gone through a process of oligarchyzation, restricting the historically high levels of internal diversity that characterized each partisan offer and deterring wider participatory processes within parties. Concurrently, the partial elimination in 1996 of the DSV for the congressional lower-FKDPEHU ZKLFK GLG DZD\ ZLWK WKH іHOHFWRUDO FRRSHUDWLYHVґ DOVR LQWURGXFHG KLJKHU FRQVWUDLQWV DW WKH WLPH RI NHHSLQJ WKH FODVVLF іHOHFWRUDO rastrLOORVґ (catch-all partisan offer) open. As both a Colorado (Foro Batllista) and a Blanco (Herrerismo) local leader state: "They have discouraged the little ants. We are little, but we are the ones gathering votes for them. Before, every weekend, I went out to the countryside, to little towns, to talk to the people. 7RGD\ , GRQ¶W GR LW DQ\PRUH. I have maintained the friendship with the people, but we cannot sacrifice friends for politics. :H FDQQRW JR RXW DQG SURPLVH ZKDW ZH GRQ¶W KDYH. :H GRQ¶W KDYH anything now. So, we set up the list with a group of friends who had a good economic situation, so we did not have to promise anything. If someone came and ask for something, we just told them that we did not have anything to give. The only thing we promised was to try to force an internal change in the party". (Hubarй Aliano, Colorado local activist. interviewed by Luna, 2003). "The political power in Montevideo is forgetting us. $QG WKDW¶V D WHUULEOH mistake. We cannot be connected only when they need us for the elections, there has to be a better way of staying in touch. 7KDW¶V WKH WUDGLWLRQ RI KHUUHULVPR WKDW¶V ZK\ ZH ZHUH VR VWURQJ. Now, national leaders have disappeared and that hurts the party. This time they did not provide political offices to us. Let¶VVD\WZRRUWKUHHRIILFHVLQWKHVWDWHDQ\WKLQJVRPHZKHUH. In the committee you have people working all year round and they are the ones keeping the presence of the party alive here. Then, when elections come they benefit from that. But when we go there, they shut the door on your face. And if the party has no reciprocity with us, we cannot provide for the people. Then, 31
how can I go and ask them for their vote?" (Juan Creceri, Blanco local activist, interviewed by Luna, 2003). Paradoxically, the weakening of party territorial structures enhanced the need of individual congress-members to stay personally in the field. This is particularly important for non-Metropolitan congress-members, which face specific constituent demands that cannot be addressed through the media. Along these lines, a Colorado congress-member points to the changes brought about by the increasing restriction of clientelistic side-payments in the system: "New opportunities had been created. You have to get to the ground from the heights and be close to the people. Talk to them, give them opinions, and inform them. Before the congress-member visit every location once a year, at most. And then, people had to go to your office, stand on a line, and present their demands to you. Some congress-member feel threatened by the LPSRVVLELOLW\ RI JHWWLQJ RQ WKH SKRQH DQG VROYH SHRSOH¶V SUREOHPV EXW WKH\ GRQ¶WUHDOL]HWKDWWKH\QHHGWRUHODWHWRSHRSOHRQQHZJURXQGV. I came from a small town and when I was a child, you rarely saw a congress-member in the field. Today, everyone is there doing different things". (Jorge Duque, Colorado congress-member, interviewed by Luna, 2003). Additionally, partially deriving from the reduction of the available stock of goods for establishing non-programmatic linkages with constituents (but also resulting from the institutional incentives introduced in the 1996 Constitution), both traditional parties have witnessed a process of bi-fractionalization, which reinforces the reduction of the historically high levels of internal competition observed in both parties (Piсeiro 2004). "Even though I am herrerista, I know that the hegemony of Herrerismo has hurted the party. We have always been a party of many candidates, three, four, or even more. And last time, we came to the national election with only one candidate and after a primary that had left many injuries open. The Blancos do not like to vote like that. They like many candidates, the Blanco Independiente and the Herrerista are different. They like different candidates and those GLIIHUHQFHV EHQHILW WKH SDUW\¶V HOHFWRUDO SHUIRUPDQFH". (Juan Creceri, Blanco local activist, interviewed by Luna, 2003). 7KHUHIRUH WKH FRQVWUDLQLQJ UXOHV LQWURGXFHG LQ KDYH UHVWULFWHG WUDGLWLRQDO SDUWLHV¶ capacity to present a widely diversified electoral menu. This has seemingly translated into a greater incapacity to compete in the long-run. 7ZRDGGLWLRQDOH[WHUQDOLWLHVі6DYLQJґ)$IURPJRYHUQLQJGXULQJWKHHFRQRPLFFULVLVDQG further enabling the consolidation of Tabarй Vбzquez political leadership. 32
The reform process and its aftermath yielded two additional negative externalities for traditional parties. On the one hand, if the reform were not approved in 1996, Frente Amplio would have probably arrived into presidential office in 1999. Accordingly, this party would have been in charge of governing the country during the most severe economic and social crisis that Uruguay had faced in history. Presumably, traditional parties would have eventually benefited from voter discontent with the left, strengthening their chance of resuming their historical leadership in 2004. Conversely, in 2004, they obtained their worst electoral result ever. On the other hand, within Frente Amplio, 9i]TXH]¶V confrontation with Seregni and Astori over the constitutional reform of 1996 triggered important degrees of internal turmoil. However, in the long-run, the internal conflict surrounding the constitutional reform debate was crucial in fostering Vбzquez hegemony and moderation. While Vбzquez opposed the reform, Seregni (still the President of FA) had personally negotiated the deal with Blanco and Colorado leaders and therefore supported the measure. Astori joined Seregni and both claimed that beyond short-term hindrances to the electoral chances of FA in 1999, the partial elimination of the DSV and the instauration of unique presidential candidacies coincided with the historical positions of FA. 6HUHJQL¶VUHVLJQDWLRQWR)$¶VSUHVLGHQF\GXULQJWKHFHOHEUDWLRQ RI WKH SDUW\¶V th anniversary in February 1996 anG 9i]TXH]¶ DVFHQW WR WKDW SRVW IXUWKHU consolidated the internal power of the second within the party. Although the reform was SRSXODUO\ UDWLILHG LQ D SOHELVFLWH 9i]TXH]¶ VWDQFH DJDLQVW WKH UHIRUP ZDV VXSSRUWHG E\ DOPRVWDERYH)$¶VYRWHLQWKH94 election) of the electorate. $GGLWLRQDOO\ LQ WKH FRQWH[W RI WKH FRQVWLWXWLRQDO UHIRUP GHEDWH 9i]TXH]¶ UHIXVDO WR adhere to an arrangement with the traditional parties consolidated his image as a harsh opposition leader. Internally, this granted Vбzquez the support of radical groups that had since 1990 aligned with Seregni and Astori, producing a crucial switch in the internal and external SRVLWLRQLQJRI)$¶VOHDGHUVKLSVDV$VWRULEHJDQWREHVHHQDVDPRGHUDWHZLOOLQJWRSDFWZLWK the traditional-family (Yaffй 2005). Thereafter, Vбzquez received further compensations for his consistent opposition to neoliberal parties with the integration of the MPP to his internal coalition in 1999 and the results he obtained in the presidential primary defeating Astori by an uncontestable margin (82% vs. 18%). /DWHUWKDW\HDU$VWRUL¶VIDFWLRQ$8VXIIHUHGGHIHDWDW the polls, obtaining almost 50% fewer votes than the largest internal plurality obtained in the 33
1994 election (39% vs. 20%). Vбzquez hegemony within FA turned out crucial in explaining WKHSDUW\¶VVXFFHVVIXOHOHFWRUDOVKRZXSLQ/DQ]DUR%XTXHW 4. Conclusion: The paradox of the reforms of 1996 In 1996 Uruguayan citizens approve a new constitution that changed several of the classic institutional features of the country since 1942. Interesting, these changes were processed in a certainly democratic environment and far away from any political, social or economic stressful situation, as seen in many other Latin American countries. The constitutional reform of 1996 is consistent with the long-term trajectory of the country. The reform introduced a crucial split in the general pattern of cooperation and competition between coalitional partners (the Blanco and Colorado party) yielding two distinct scenarios (from 1984 to 1995 and since 1996) that potentially introduce important discontinuities for the analysis of political competition in the country. In the 1994 election the leftist party (Frente Amplio) almost won the presidency. This fact marked a fundamental ZDWHUVKHGRQWKHFRXQWU\¶VUHFHQWKLVWRU\LQFUHDVLQJWKHSHUFHLYHGFRVWVRIORRVLQJRIILFHIRU both traditional parties. While before 1994, losing office meant that the other traditional party would win but without a majority, now the implications were far more drastic as both traditional parties would be left out of office. :KLFKIDFWRUVH[SODLQFRQVWLWXWLRQDOUHIRUPHUV¶P\RSLDLQ8UXJXD\")ROORZLQJ3LHUVRQ (2004) we will claim that this case illustrates a clear instance in which the instrumental and short-term orientation of reformers contradicted their long-term interests. Therefore, in Uruguay, reformers were short-sighted and unable to anticipate the long-term effects of some measures that were introduced jointly with the runoff system to successfully achieve their short-term objective (prevent FA to win the election of 1999). However, this case also illustrates the limitations that instrumental reformers face when they need to negotiate an incoherent reform package to accommodate the interest of a wide set of partisan actors and to make that package attractive to the citizenry. Finally, we also need to stress that the long-term outcomes of the reform (i.e. the result of the 2004 election) were not only a by-product of its long-term effects. Indeed, that outcome was also a consequence of a long-run electoral swing and facilitated by the occurrence of a very significant external shock: the financial collapse of the country in 2002. Although some of the external phenomena affecting the long-term 34
effectiveness of the reform should have been predicted (i.e. the long-term pattern of electoral growth of the left), the depth of the economic crisis and its effects on partisan alignments when interacting with the externalities created by the Constitutional reform were not foreseeable. Given that any reform of the constitution must be approved by a majority of the electorate in Uruguay, the reform should not only be functional to the interest of both traditional parties, but should also seek to accommodate broader interest in order to generate a critical mass of partisan support for the reform. Moreover, facing the challenge of making the reform legitimate and worthy in the eyes of the citizenry, interest should be dressed with reasons. The reasons on which citizen support for the reform was requested derived from some of the anticipated consequences of the four amendments address above. Whereas our evidence shows that the first reason on which reform was advocated was unfunded (coalitional politics was further complicated after the reform was introduced), we claim that the second series of reasons XVHG WR іVHOOґ WKH UHIRUP WR WKH SXEOLF ZDV PRUH accurately predicted and promised. Precisely, the effects of the additional reforms introduced to accommodate the interests of a broad partisan coalition and to lower the prominence of the runoff system within the reform package, ended up producing more harm than good to traditional parties in the long-run. Especially the single presidential candidacies and the HOLPLQDWLRQ RI µHOHFWRUDO FRRSHUDWLYHV¶ EDVHG RQ WKH '69 HURGHG WKH HOHFWRUDO EDVHV RI WUDGLWLRQDOSDUWLHVDQGWKHLUFDSDFLW\WRRIIHUDZLGHіHOHFWRUDOPHQXґ 0RUHRYHUGXHWRWKH institutionalization of a bi-polar pattern of political competition between the left and traditional parties which further complicated coalitional politics and made the latter increasingly unable to differentiate and renew their electoral offer. Uruguay excels in Latin America for having tried with an enormity of political institutions, most of them processed within a democratic context. The key actors for all these experiences have being political parties and one of their most outstanding features is the existence of strongly organized fractions that operate with great political autonomy inside them. The new rules of the game approved in 1996, while fostering more fragmented political configurations than those Uruguayan used to witness, did not give the government effective instruments which would serve to neutralize this fragmentation. The only arena in which the new rules could have some impact is in the legislative discipline of fractions and parties, a 35
phenomenon which, in any case, was present before the reform (Buquet, Chasquetti, and Moraes 1998). 36
Figure 5. Ballot paper for the 2004 national elections 37
5. References Altman, David. 2000. "The Politics of Coalition Formation and Survival in Multiparty Presidential Democracies: The Case of Uruguay 1989-1999." Party Politics 6 (3): 259-283. Altman, David. 2002. "Popular Initiatives in Uruguay: Confidence Votes on Government or Political Loyalties?" Electoral Studies 21 (4): 617-630. Altman, David. 2008. "Collegiate Executives and Direct Democracy in Switzerland and Uruguay: Similar Institutions, Opposite Political Goals, Distinct Results." Swiss Political Science Review 14 (3): 483-520. Altman, David and Rossana Castiglioni. 2006. "The 2004 Uruguayan Elections: A Political Earthquake Foretold." Electoral Studies 25 (1): 147-154. Altman, David and Daniel Chasquetti. 2005. "Re-Election and Political Career Paths in the Uruguayan Congress: 1985-1999." The Journal of Legislative Studies 11 (2): 235-253. Atkeson, Lonna Rae. 1998. "Divisive Primaries and General Election Outcomes: Another Look at Presidential Campaigns." American Journal of Political Science 42 (1): 256271. Buquet, Daniel. 2001. "Partidos, Elecciones y Gobernabilidad en Uruguay: Los Posibles Efectos de la Reforma Constitucional de 1996." In Transiciуn Democrбtica y Gobernabilidad: Mйxico y Amйrica Latina, edited by Julio Labastida, Antonio Camou, and N. Lujбn. Mйxico: IIS-UNAM, Flacso, Plaza y Valdйs Editores, Pp. Buquet, Daniel. 2003. "El Doble Voto Simultбneo." Revista de la Sociedad Argentina de Anбlisis Polнtico 1 (2). Buquet, Daniel. 2004. "Balotaje vs. Mayorнa Simple: el "Experimento2 Uruguayo." In La elecciуn presidencial mediante doble vuelta en Latinoamйrica, edited by R. Martнnez Mбrtнnez. Barcelona: Institut de Ciencies Polнtiques i Socials. Universitat Autтnoma de Barcelona, Pp. 483-500. Buquet, Daniel (Ed.). 2005. Las Claves del Cambio: Ciclo electoral y nuevo gobierno 2004/2005. Montevideo: Banda Oriental - Instituto de Ciencia Polнtica. 38
Buquet, Daniel and Rafael Piсeiro. 2000. "Coaliciуn de Gobierno y Reforma Constitucional: Un Enlace Estratйgico." In Coparticipaciуn y Coaliciуn: 164 Aсos de Acuerdo entre Blancos y Colorados. Montevideo: Arca-Humus, Pp. Buquet, Daniel, Daniel Chasquetti, and Juan Andrйs Moraes. 1998. Fragmentaciуn Polнtica y Gobierno en Uruguay: їUn Enfermo Imaginario? Montevideo: Facultad de Ciencias Sociales. Castiglioni, Rossana. 2005. The Politics of Social Policy Change in Chile and Uruguay: Retrenchment versus Maintenance 1973-1998. New York & London: Routledge. Colomer, Josep M. 2002. "Las Elecciones Primarias Presidenciales en Amйrica Latina y sus Consecuencias Polнticas." In El Asedio a la Polнtica, edited by Marcelo Cavarozzi and J. M. A. Medina. Rosario: Homo Sapiens, Pp. Coppedge, Michael. 1994. Strong Parties and Lame Ducks: Presidential Partyarchy and Factionalism in Venezuela. Stanford: Stanford University Press. Espнndola, Roberto. 2001. "No Change in Uruguay: The 1999 Presidential and Parliamentary Elections." Electoral Studies 20 (4): 649-657. Filgueira, Carlos and Fernando Filgueira. 1997. "Coaliciones Reticentes: Sistema Electoral, Partidos y Reforma en el Uruguay de Fin de Siglo." Cuadernos del CLAEH 78/79, 325-347. Finch, Henry. 1995. "The Uruguayan Election of 1994." Electoral Studies 14 (2): 232-236. Gonzбlez, Luis E. 1995. "Continuity and Change in the Uruguayan Party System." In Building Democratic Institutions: Party Systems in Latin America, edited by S. Mainwaring and T. Scully. Stanford: Stanford University Press, Pp. 138-163. Gonzбlez, Luis Eduardo. 1993. Estructuras Polнticas y Democracia en Uruguay. Montevideo: Instituto de Ciencia Polнtica. Gros Espiell, Hйctor. 2002. "Uruguay: їPresidencialismo o Parlamentarismo?" Cuestiones Constitucionales 7 (Julio-Diciembre): 87-108. Guerrini, Aldo. 2000. "El peso politico de los intendentes del interior: apuntes para una discusion a la luz de las ultimas consultas electorales." In Elecciones 1999/2000, edited by G. Caetano. Montevideo: Ediciones de la Banda Oriental - Instituto de Ciencia Polнtica, Pp. 185-193. 39
Jones, Mark P. 1995. Electoral Laws and the Survival of Presidential Democracies. Notre Dame and London: University of Notre Dame Press. Kenney, Patrick J. and Tom W. Rice. 1987. "The Relationship between Divisive Primaries and General Election Outcomes." American Journal of Political Science 31 (1): 31-44. Lanzaro, Jorge (Ed.). 2004. La Izquierda Uruguaya: Entra la Oposiciуn y el Gobierno. Montevideo: Instituto de Ciencia Polнtica. Lindahl, Gцran. 1977. Batlle y la Segunda Constituciуn, 1919-1933. Montevideo: Arca. Magri, Altair. 2000. "La llave del triunfo electoral en Mayo 2000: Reelecciуn, continuidad y 'por la vuelta'." In Elecciones 1999/2000, edited by G. Caetano. Montevideo: Ediciones de la Banda Oriental - Instituto de Ciencia Polнtica, Pp. 165-183. Mainwaring, Scott and Matthew Shugart (Ed.). 1997a. Presidentialism and Democracy in Latin America. Cambridge: Cambridge University Press. Mainwaring, Scott and Matthew S. Shugart. 1997b. "Conclusion: Presidentialism and the Party System." In Presidentialism and Democracy in Latin America, edited by S. Mainwaring and M. S. Shugart. Cambridge: Cambridge University Press, Pp. 394-439. Mieres, Pablo. 1992. "Acerca de los cambios del sistema de partido uruguayo." Cuadernos del CLAEH 62, 65-75. Pierson, Paul. 2004. Politics in Time. History, Institutions, and Social Analysis. Princeton and Oxford: Princeton University Press. Sacchi, Martнn. 1999. "Partidos, Fracciones y Gobierno en el Segundo Colegiado (19521966)." Revista Uruguaya de Ciencia Polнtica 11, 9-34. Sartori, Giovanni. 1976. Parties and Party System. Cambridge: Cambridge University Press. Shugart, Matthew and John Carey. 1992. Presidents and Assemblies: Constitutional Design and Electoral Dynamics. Cambridge: Cambridge University Press. 40

File: constitutional-reforms-and-political-turnover-in-uruguay-winning.pdf
Title: AltmanEtAl_DOL(1).pdf
Published: Thu Jun 30 18:56:25 2011
Pages: 41
File size: 0.89 Mb

WE ARE WHAT WE EAT, 30 pages, 1.05 Mb
Copyright © 2018