Abstract
This article employs roll call vote data and Bayesian ideal point estimation to examine inter-party dynamics in the Argentine Chamber of Deputies between 1989 and 2007. It highlights the presence in the Argentine Congress of a strong government vs. opposition dimension as well as identifies the relative position on this dimension, vis-à-vis the governing party, of the most prominent non-governing parties. Special attention is paid to the evolution of inter-party legislative dynamics during Argentina's brief experience with coalition government (1999-2001) and to party behavior in the Chamber during the final two years of President Néstor Kirchner's term in office (2005-07).
Introduction
Over the past twenty years Argentina has experienced two economic melt-downs and several dramatic leadership changes. 1 The economic and social disorder of the final year of the tenure of President Raúl Alfonsín (1983-89) of the Unión Cívica Radical (UCR) was followed by ten years of relative stability and a wide scale implementation of neoliberal policies under President Carlos Menem (1989-99) of the Partido Justicialista (PJ). In part as a reaction to problems of poor governance quality and corruption during the Menem Administration, in 1999 President Fernando de la Rúa of the UCR was elected as part of a UCR-FREPASO (Frente País Solidario) alliance (the Alianza), only to be forced out of power in late 2001 as Argentina suffered a severe economic, social, and political crisis. By 2003, the worst of the crisis had passed, and Argentina began to experience renewed economic growth and stability under the leadership of Presidents Eduardo Duhalde (2002-03) and Néstor Kirchner (2003-07) of the PJ. Under Kirchner, Argentina reversed many of the neoliberal policies of the 1990s and replaced them with a much more prominent role for the state in the Argentine economy. 2 Following the underlying logic of Peronism, most of the PJ politicians who had been fervent advocates of neoliberal policies under President Menem became fervent advocates of statist policies under President Kirchner.
We thank Silvina Danesi, Alberto Föhrig, Jorge Gordin, Marcelo Leiras, one anonymous reviewer, and seminar participants at the Universidad de San Andrés (Buenos Aires, Argentina) for their helpful comments and suggestions.
At the end of the presidency of Carlos Menem, Argentina had a Heritage Foundation Index of Economic Freedom score of 71, which ranked 20th among the nations of the world. At the end of the presidency of Néstor Kirchner, Argentina had an Index of Economic Freedom score of 55, which ranked 109th among the nations of the world (Heritage Foundation 2008).
In this article we employ roll call vote data to examine inter-party dynamics in the Argentine Chamber of Deputies between 1989 and 2007. The data highlight the strong government vs. opposition dimension present in the Argentine Congress as well as the relative position on this dimension of the most relevant non-governing parties vis-à-vis the governing party during this time period.
Section 2 provides a brief summary of Argentine political institutions, with a particular focus on the Chamber of Deputies. Section 3 reviews the Argentine political party system and discusses the country's most relevant political parties. Section 4 describes the roll call vote data examined here as well as summarizes some general conclusions regarding the government-opposition dimension that dominates roll call vote behavior in the Argentine Congress. Section 5 analyzes inter-party dynamics in the Chamber of Deputies, detailing the relative location of the key partisan actors on the government-opposition continuum. Section 6 deepens the previous discussion through an in-depth analysis of the inter-party dynamics between the main partners of the ill-fated Alianza (UCR, FREPASO) that governed Argentina between 1999 and 2001. Section 7 complements the preceding analysis with an exploration of inter-party dynamics during the more recent legislative period (2005-07), highlighting the location of the most relevant legislative delegations on the government-opposition dimension during this time frame. Section 8 concludes.
The Argentine president is constitutionally quite powerful, possessing both a strong veto as well as executive decree authority (Benton 2002; Corrales 2002; Negretto 2004). Argentina has a bicameral national legislature and a federal system of government in which provincial governors exercise substantial autonomy (Gordin 2004; Remmer and Wibbels 2000; Spiller and Tommasi 2007). Every one of the 24 provinces (23 provinces and an autonomous federal capital) possesses three senators (two prior to 1995) and a number of deputies proportional to its population in 1980, with every province receiving a minimum of five deputies. The Argentine Congress, while certainly much more of a reactive blunt veto player than a proactive agenda setter, is nevertheless an important actor in the policy process (Jones, Saiegh, Spiller, and Tommasi 2002; Llanos 2002; Mustapic 2002).
During the 1989-2007 period the majority party enjoyed considerable success within the Chamber of Deputies. This success stems from the majority party's ability to function as an effective cartel (Calvo 2007; Jones and Hwang 2005; Jones and Hwang 2006). The majority party leadership uses its majority status (especially agenda control) to dominate the legislative process, excluding legislation it believes may pass despite its objection (negative agenda control), as well as implementing legislation it desires (positive agenda control). The opposition, lacking both negative and positive agenda control, is left in a very reactive position. As a result, the best way to interpret roll call vote behavior in Argentina is as the legislator's position (for or against) on the legislation placed on the floor agenda by the majority party.
The Argentine Political Party System
Between 1989 and 2007, Argentine politics was dominated by two political parties, the Partido Justicialista (PJ, Peronists) and, to a diminishing extent as time passed, the Unión Cívica Radical (UCR, Radicals). Periodically, another party achieved a modest degree of national prominence. 3 During the 1989-2007 period, the four parties that achieved a noteworthy level of seats in the Chamber of Deputies are the Unión del Centro Democrático (UCEDE), Frente País Solidario (FREPASO), Acción por la República (AR), and Afirmación para una República Igualitaria (ARI). While the UCEDE and AR are generally identified with the center-right of the political spectrum, FREPASO and ARI are normally located on the center-left (Alcántara Sáez 2008; Escudero 2001; Gibson 1996). To date, these parties have seen their electoral support evaporate after only a few elections, with only the ARI (founded in 2000) still a viable political force as of 2007. In contrast to the PJ and UCR, these minor national parties never established an effective party organization and overly depended on the popularity of a single leader (or small coterie of leaders). Furthermore, these parties have consistently failed to significantly branch beyond their initial core geographic area of support, the Capital Federal and the portion of the Province of Buenos Aires adjacent to the Capital Federal (Greater Buenos Aires) (Malamud and De Luca 2005). 4
The 2003-07 period witnessed a modest increase in the degree of splintering within the PJ, with several prominent neo-PJ delegations established in the Chamber such as the Frente del Movimiento Popular (FMP), Justicialista Nacional (JN), and Peronista Federal (PF). For the most part these delegations only survived for one legislative period prior to returning either to the PJ-fold or merging with/forming another delegation.
Some of these parties have nonetheless experienced modest, albeit sporadic, success in a small number of interior provinces in a handful of elections (e.g., the ARI in Santa Fe and Tierra del Fuego and the FREPASO in Neuquén and Santa Fe).
Argentina has a large number of parties that compete either solely (or effectively) in only one province. The members of this diverse family of parties are collectively referred to as “provincial parties,” and are generally considered to be located on the center-right of the political spectrum given either their origins, base of support, or both (Gibson 1996). 5 Here we examine the four most prominent provincial parties during the 1989-2007 period: the Fuerza Republicana (FR) of Tucumán, the Movimiento Popular Neuquino (MPN), the Partido Demócrata de Mendoza (PD), and the Partido Renovador de Salta (PRS). 6
This center-right characterization applies best to the majority of these parties that trace their origins to Argentina's oligarchic period (PD), were founded by a former military governor who had held office during the 1976-83 military dictatorship (FR, PRS), and/or collaborated actively with the military government during the 1976-83 dictatorship (PD, PDP). It applies less well for some other parties (MPN, in spite of some collaboration).
These are the four parties with the highest median number of deputies (where there was a tie, the party with the highest mean number of deputies during this period was selected) during the ten legislative periods contained within the 1989-2007 time frame. It should be noted that these median seat numbers are quite small, ranging from one to two. Also included in some of the later analysis of the provincial parties is the Partido Demócrata Progresista (PDP), since if the PDP were considered a provincial party, it would displace the PD as one of the four most prominent provincial parties during this period. However, while most PDP deputies between 1989 and 2007 came from its bastion in the province of Santa Fe, during four legislative periods one PDP deputy represented the Capital Federal. The FR also is technically a national party, although all of its electoral success occurred within its home province of Tucumán.
The PJ-UCR dominance can be seen in the substantial control exercised by these two parties (especially the PJ) over the most important political posts in the country: the presidency, in Congress (Senate and Chamber of Deputies), the governorships, and the mayoralties (Jones and Hwang 2006; Malamud and De Luca 2005). For instance, the 1989-2007 period comprises the two terms of President Carlos Menem (1989-95, 1995-99) of the PJ, the abbreviated tenure of President Fernando de la Rúa (1999-2001) of the UCR, the period of interim president Eduardo Duhalde (2002-03) of the PJ, and the term of President Néstor Kirchner (2003-07) of the PJ. 7
Between 1983 and 1989, Raúl Alfonsín (UCR) occupied the presidency.
Table 1 provides additional information on the dominance of the PJ and UCR since the return to democracy in 1983. The PJ in particular has consistently held a large percentage of seats in the Chamber of Deputies, ranging from 39 percent to 55 percent during the 1989-2007 period. The UCR witnessed a severe decline in its electoral fortunes over the past 20 years, and during the 2005-07 period occupied a mere 16 percent of the seats in the Chamber. Nonetheless, between 1989 and 2007 the UCR always possessed the second largest delegation in the Chamber, albeit with the gap between it and the next largest party dropping from a high of 31 percent in 1991 to a low of 8 percent in 2005. 8
At the start of the 2007-09 legislative period, the UCR delegation had fallen to an anemic 9 percent of the Chamber seats. However, reflecting the extreme fragmentation of the non-governing parties in the Chamber, the UCR still remained the second largest delegation (albeit by only 2 percent).
The Composition of the Argentine Chamber of Deputies, 1983-2007
Note: All parties that in at least one two-year legislative period held 3 percent or more of the seats are listed separately above. All Provincial Parties are grouped together.
Source: Authors’ elaboration based on data from Jones and Micozzi (2009).
Table 1 also details the rise and decline of Argentina's minor national parties, with parties that at one time represented the third or fourth largest delegation in the Chamber, either not occupying any seats (UCEDE, FREPASO) or a forlorn seat (AR) in 2005. Only the ARI, the youngest of these parties, continued to possess a viable legislative delegation as of 2005.
Roll call votes provide an unparalleled empirical source of information on legislators and political parties. As Krehbiel (1998) notes, roll call vote data are imperfect, given the fact that they occur at the very end of the legislative process. Nonetheless, there is a tremendous amount of useful information that can be obtained from the analysis of roll call vote data (Clinton, Jackman, and Rivers 2004; Poole 2005). The use of legislator ideal points to study congressional behavior is the norm in U.S. legislative studies (e.g., Aldrich, Berger, and Rohde 2002; Poole 2005; Poole and Rosenthal 2007) and is increasingly popular for the study of legislatures in other countries (e.g., Alemán and Saiegh 2007; Hix, Noury, and Roland 2006; Morgenstern 2004).
The Argentine Chamber of Deputies possesses an electronic voting system (overhauled in 2001) that is used for both recorded (roll call) and unrecorded votes. A roll call vote is taken when a motion for one is made and then supported by at least one-fifth (1989-96) or one-tenth (1997-2007) of the deputies in attendance. 9 In addition, since June of 2006 all general votes on proposed laws (i.e., on the entire bill, prior to any article-by-article votes) are automatically recorded.
All roll call vote data and information on session attendance and delegation affiliation were obtained from the Argentine Chamber of Deputies. Only 39 roll call votes were held between December 10, 1983 and July 7, 1989 (Jones 2002). For additional information on roll call vote procedures and data in Argentina, see Jones (2002), Jones and Hwang (2005), and Morgenstern (2004).
This article employs roll call votes taken between July 8, 1989 (from the date President Menem assumed office and the PJ assumed de facto control of the Chamber of Deputies) and December 9, 2007. 10 These data represent ten full two-year legislative periods, corresponding to the biennial partial renovations of the Chamber (1989-91, 1991-93, 1993-95, 1995-97, 1997-99, 1999-2001), and one partial period (July 8, 1989 to December 9, 1989). In the analysis, two of these legislative periods are split. In the first case, the 1999-2001 legislative period is divided into one period prior to Vice President Carlos “Chacho” Álvarez's (FREPASO) resignation on October 6, 2000 (December 10, 1999 to October 5, 2000) and a second period following this resignation (October 6, 2000 to December 9, 2001). The 2001-03 period is divided between the administration of interim president Eduardo Duhalde (January 2, 2002 to May 24, 2003) and that of President Néstor Kirchner (May 25, 2003 to December 9, 2003). 11
Legislative periods begin on December 10 of odd years and end on December 9 of the following odd year. In 1989, the severe crisis facing the country resulted in an agreement between outgoing President Raúl Alfonsín and incoming President Carlos Menem for the latter to assume office early on July 8, 1989, instead of on the constitutionally mandated date of December 10, 1989 (Jones 1997).
These legislative periods (1999-2001 and 2001-03) are each split into two different periods due to the extraordinary changes that occurred during them (i.e., the severe decay of the UCR-FREPASO Alianza coalition following Álvarez's resignation for the former period and the change in the control of the presidency from Duhalde to Kirchner in the latter period). These extraordinary changes make it undesirable to examine each period as a single unit (as is done for the other periods analyzed here).
Included in the initial analysis population are all valid roll call votes held during the legislative period. Due to problems of convergence stemming from unanimous or near-unanimous votes, following the logic of Poole and Rosenthal (2007) we exclude all lopsided roll call votes in which fewer than 3 percent of the deputies were on the losing side. 12
An exception was made if a binary logit analysis of large party (PJ, UCR, FREPASO [1995-2001]) deputies detected a significant relationship between party affiliation and a deputy's strategic absence from the floor (or abstention) at the time of voting. In those cases where a significant difference was detected, the roll call vote was retained in the final analysis population.
Jones and Hwang (2005) employ a Bayesian estimation procedure to uncover the number of dimensions present in Argentine roll call vote behavior (they find a single dimension underlies this voting behavior), and then to identify the Chamber deputies’ median ideal points (i.e., the location of each deputy's preferred location on this single dimension). Each deputy's ideal point (ranging from −1.0 to 1.0) represents the median of 10,000 (or 5,000) draws from the deputy's posterior distribution of ideal points. 13
Only deputies present for more than 20 percent of the roll call votes during a legislative period are examined in the following analysis.
Table 2 provides details on our analysis of these roll call vote data. 14 Two points are of principal note. First, with the exception of the 2005-07 period, 15 the number of valid roll call votes varies across the different periods of analysis from lows in the twenties to highs of slightly more than one hundred. Due principally to a change in the Chamber's internal rules regarding the use of roll call votes, there were more roll call votes taken between 2005 and 2007 than during the entire 1989-2003 period combined. Second, virtually all of the roll call votes (mean of 95 percent, median of 96 percent) in the Argentine Chamber discriminate on a single dimension. The results of the roll call vote analysis presented in Table 2 demonstrate quite clearly that the roll call votes in the Argentine Chamber can be successfully accounted for by a one-dimensional spatial model.
For reasons of space we do not provide a complete description of the statistical methodology employed for our analysis. For a more complete coverage of the methodology employed here to analyze these Argentine roll call votes via Bayesian simulation, see Jones and Hwang (2005).
The results of the analysis for 2005-07 (both those presented here and those not included for reasons of space) do not differ in a noteworthy manner from those for the pre-2005 legislative periods when significantly fewer roll call votes were taken. A comparative analysis of the pre and post 2005 periods suggests our findings for the entire 1989-2007 time period are representative of general floor legislative dynamics, and not, for instance, biased by the small sample of legislative bills on which roll call votes were taken between 1989 and 2005. Related confirmation of the robustness of these findings is provided by Alemán, Calvo, Jones, and Kaplan (2009) who compare roll call vote based ideal points with ideal points derived from bill cosponsorship in their study of the Argentine Chamber.
One Dimensional Analysis of Roll Call Data via Bayesian Simulation
Note: Beta ~ N(0,0.5). Burn-in before saving last 10,000 obs. (in some cases 5,000 obs.*).
Source: Authors’ elaboration based on data from Jones and Micozzi (2009).
This single dimension can be described in several ways; however, the description that is most accurate is one that views this continuum in terms of a government vs. opposition cleavage (Jones and Hwang 2006). 16 Under this interpretation, legislative voting separates deputies into government and opposition camps based on their partisan affiliation, with the governing party (also the majority in the Chamber) functioning as a cartel that effectively controls the agenda in both a negative and positive manner, thereby reducing the ability of the opposition to influence policy (especially in a positive manner) (Cox and McCubbins 2005).
This single dimension is not primarily ideological, as is the case in the United States and Western Europe (Poole and Rosenthal 2001).
This section analyzes two key aspects of the location of political parties on the government-opposition dimension along which voting occurs in the Argentine Chamber of Deputies. The evolution of median party ideal points during the 1989-2007 period as well as the level of major party overlap is first examined. Then, in order to allow for greater inter-temporal comparison, the political parties are located on the government-opposition dimension for each legislative period relative to identifiable baselines: the governing party and the principal opposition party.
Median Party Ideal Point Evolution and Major Party Overlap
Figure 1 charts the evolution of the median ideal point of the PJ, UCR, and the four most relevant minor national parties active during the 1989-2007 period (UCEDE, FREPASO, AR, ARI). 17 The PJ and UCR represent the two extremes of this dimension for most of this period. The PJ held the presidency and controlled the Chamber throughout the 1989 to 2007 span, with the exception of the 1999-2000 and 2000-01 periods during which time the UCR held the presidency and a UCR-FREPASO alliance (i.e., the Alianza) controlled the Chamber. The distance between the PJ and UCR median ideal points remained large and relatively constant throughout this time period, except for 2001-03 and 2003 when the UCR moved towards the center, with a less oppositional relationship with the governing PJ than was the case in other legislative periods. This temporary closer proximity stemmed largely from the severe economic, social, and political crisis experienced by Argentina in 2002 and 2003, and the UCR decision to support the PJ government's legislative agenda to a much greater extent than was normally the case. By the 2003-05 period, when the worst of the crisis had passed, the UCR returned to its more traditional oppositional position vis-à-vis the governing PJ.
The ideal points for the minor national parties are only provided for legislative periods in which their respective delegation had four or more members.
Government vs. Opposition I: PJ, UCR, and Minor National Parties, 1989-2007
Figure 2 supplies comparable information to that in Figure 1 for the PJ, UCR, and four principal provincial parties (FR, MPN, PD, PRS). With a few exceptions, the provincial parties tended to occupy a middle ground between the governing party and principal opposition party (i.e., between the PJ and UCR).
Government vs. Opposition II: PJ, UCR, and Principal Provincial Parties, 1989-2007
Overlap Between the PJ and UCR
Three principal conclusions can be drawn from Figures 1, 2, and 3 regarding the 1989-2007 period. First, focusing on the two dominant parties (the PJ and UCR), the level of inter-party heterogeneity in roll call vote behavior remained both high and relatively constant (with the partial exception of the UCR's shift during the 2001-03 and 2003 periods). There also was little overlap between the ideal points of the PJ and UCR deputies. In sum, there is little variation in the level of major party inter-party heterogeneity except, to a modest extent, for the crisis years of 2001 to 2003.
Second, the behavior of the four minor national parties is illustrative of the government-opposition nature of this single dimension upon which roll call voting takes place in the Argentine Chamber. In three of the four instances, these parties of the center-right (UCEDE, AR) and center-left (FREPASO) on average occupied a middle ground between the PJ and UCR on this government-opposition dimension, at times voting with the government and at other times with the opposition. The UCEDE and AR in particular spent either all (AR) or most (UCEDE, all years but 1993-95) of these legislative periods at a location on this dimension between that of the governing party and the principal non-governing party.
Third, the ARI's experience most directly highlights the government-opposition nature of this dimension. The ARI was founded by former FREPASO and UCR deputies under the leadership of Elisa Carrió (ex-UCR) in 2000. From the outset the ARI presented a firm opposition to the government of President Fernando de la Rúa (UCR), with a median ideal point that was very similar to that of the principal opposition party (i.e., the PJ). However, with the return of the PJ to the presidency (Eduardo Duhalde [] and later Néstor Kirchner []) and control of the Chamber, the ARI's median ideal point jumped over that of the PJ and UCR, going from a 0.75 to a −0.86, that is crossing the entire dimension! The ARI went from being the most ardent opponent of the governing UCR and indistinguishable from the main opposition PJ to being the most ardent opponent of the governing PJ and noticeably more distant from the governing PJ than the principal opposition party (the UCR). The principal explanation for this change in the position of ARI on this dimension was the switch in control of the presidency and Chamber from the UCR to the PJ.
Party Location on the Government-Opposition Dimension over Time
Given the inability to directly compare the specific party median ideal point estimates across legislative periods, it is useful to provide a comparison of a party's location in each legislative period relative to identifiable baselines. Table 3 provides a summary of the location of the ten parties examined in Figures 1 and 2. It also includes the PDP (see note 6) and five parties that possessed important delegations during some or all of the 2003-07 period: the three PJ splinter parties discussed previously (FMP, JN, PF), the Partido Socialista (PS), and the Propuesta Republicana (PRO). 18
Between 1995 and 2001 PS deputies formed part of the FREPASO (and later ARI) delegations. The PRO is a party based in the Capital Federal that was founded by Mauricio Macri, Chief of Government/Mayor of the Capital Federal (City of Buenos Aires) since 2007.
Government and Opposition Blocs in the Argentine Chamber, 1989-2007
Source: Authors’ elaboration based on data from Jones and Micozzi (2009).
In Table 3, for each of the twelve legislative periods, two poles (one Government, one Opposition) were created. The former pole is anchored by the governing party, which controlled the presidency of the nation as well as the presidency of the Chamber of Deputies (the PJ in all periods except for 1999-2000 and 2000-01, when the UCR occupied both posts). The latter pole is anchored by the principal non-governing party, which is the party other than the governing party with the most seats in the Chamber (the UCR in all periods except for 1999-2000 and 2000-01, when the PJ held this status). For each of the two parties which respectively anchor the government and opposition poles, the 95 percent confidence interval for the party's median deputy also was determined.
Any party whose median deputy's 95 percent confidence interval overlaps with that of the governing party is placed in the “Government Bloc” column. Any party whose median deputy's 95 percent confidence interval overlaps with that of the principal non-governing party is placed in the “Opposition Bloc” column. Any party whose median deputy's ideal point is between that of the governing party and that of the principal non-governing party, but whose 95 percent confidence interval is not overlapped by that of either of these two anchor parties, is placed in the “Center Bloc.” Finally, where a party's median ideal point is between that of the principal non-governing party and the endpoint of the opposition side of the dimension, but whose median deputy's 95 percent confidence interval is not overlapped by that of the principal non-governing party, the party is placed in the “Ultra Opposition Bloc.” 19 The anchor parties (the PJ and UCR for all legislative periods) are shown in a row by themselves in bold, parties generally located on the center-right of the political spectrum are in italics (Alcántara Sáez 2008; Escudero 2001; Gibson 1996), parties generally located on the center-left of the political spectrum are underlined (Alcántara 2008; Escudero 2001), and Peronist splinter parties are unadorned. Parties occupying at least 3 percent of the Chamber seats during that particular legislative period are in bold.
No parties are located in the theoretically possible, but empirically unrealized, category of “Ultra Government Bloc.”
Six main points can be drawn from a review of Table 3. (1) With a few exceptions, the governing party generally rules/votes alone, with either no other party or at most a handful of minor parties present in the Government Bloc. In five of the 12 periods, only the governing party was located in the Government Bloc column. In three other periods, only a single party (holding between two and three seats) shared this classification with the governing party, while in a fourth period (1997-99) the PJ was joined by two provincial parties (each with two seats)
There are three partial exceptions to this general rule. First, there is the abbreviated 1989 period when President Carlos Menem's PJ was joined by three provincial parties and one minor national party in the Government Bloc (though combined, these parties accounted for only 11 seats). Second, during the peak of Argentina's economic, political, and social crisis in 2002 and 2003, the AR, FR, and MPN (which combined occupied a mere eight seats) were aligned in their voting behavior with President Eduardo Duhalde's PJ. Finally, there is the 1999-2000 period when the UCR, its coalition partner FREPASO (with 37 seats), and the MPN (with two seats) all were located in the Government Bloc.
(2) No party was located in the Center Bloc in half of the 12 periods. In the six other periods, in two periods only one party (the MPN, with two or three seats) was located in the Center Bloc, while in two other periods a scant two parties were present in this column. The party most often present in this space was the PRS, in 1991-93 during the government of President Carlos Menem, in 1999-2000 and 2000-01 during the government of President Fernando de la Rúa, and in 2005-07 during the government of President Néstor Kirchner. 20 The only period in which a plurality of the parties was located in the Center Bloc was 2000-01, when four parties occupied this space. FREPASO had moved here from the Government Bloc (where it was located in 1999-2000) as it distanced itself from the De la Rúa government (see Section 6), the PRS maintained its partial allegiance to the De la Rúa government, and the AR switched from the Opposition Bloc to the Center Bloc as its leader, Domingo Cavallo, joined De la Rúa's cabinet as the Minister of Economy in April of 2000.
In 1995, the PRS backed the candidacy of President Menem (PJ). The PRS presented a joint legislative list with the UCR and FREPASO in Salta in 1999, supporting the De la Rúa (UCR) presidential candidacy. In 2007 the PRS backed the candidacy of Cristina Fernández de Kirchner (PJ), sharing a joint legislative list with one of Salta's two competing Peronist factions.
(3) By far the most common location for parties during this period was in the Opposition Bloc. During all periods but 1989, the Alianza government periods of 1999-2000 and 2000-01, and the 2001-03 period, an absolute majority of the parties were located in the Opposition Bloc. In sum, when presented with the governing party's legislative agenda, most parties adopted a voting position on the floor that more commonly placed them in the Opposition Bloc than in either the Government Bloc or the Center Bloc.
(4) The Ultra Opposition Bloc was rarely occupied by any political party. Between 1989 and 2007, only in three periods was a party located in this column. The first parties to reside in the Ultra Opposition Bloc were the ARI and the PS in 2001-03, when the UCR moved away from its traditional strong opposition position and was more supportive of the governing PJ's legislative agenda than was normally the case, as a consequence of the extraordinary crisis facing Argentina at this time. In 2003-05 the ARI returned to this category due to its high level of dissent with the PJ controlled legislative agenda. In this period the ARI was joined by the PDP's single deputy.
(5) Table 3 underscores the tendency of many of the minor parties to occupy a location on this government-opposition dimension that is on average in the center. For instance, of the 12 legislative periods, the MPN could be found in the Government Bloc in four, the Center Bloc in two, and the Opposition Bloc in six. The PRS was in the Government Bloc on three occasions, the Center Bloc on four occasions, and the Opposition Bloc on five occasions. Similarly, in the four periods for which data were examined for it, the AR was in the Government Bloc once, the Center Bloc once, and the Opposition Bloc twice. In all, the mean location of all of the minor national parties and provincial parties is between that of the governing party and principal non-governing party, with one major and one minor exception. The major exception is the ARI, which since its formation in 2000 has occupied on average a location that is further to the opposition end of the dimension than even the anchor party of the Opposition Bloc. The minor exception is the PS, which on two instances was in the Ultra Opposition Bloc alongside the ARI, in one instance in the Opposition Bloc, and most recently in the Center Bloc.
(6) The conclusions reached by the analysis above are reinforced by an examination of the same five dynamics restricted to parties that during a legislative period held at least 3 percent (eight) of the seats in the Chamber (in bold in Table 3). In addition to the PJ and UCR, which enjoyed this status in all periods, seven parties held at least 3 percent of the seats during at least one of the twelve legislative periods examined: the FREPASO during five periods (1995-97, 1997-99, 1999-2000, 2000-01, 2001-03), the ARI during five periods (2000-01, 2001-03, 2003, 2003-05, 2005-07), the UCEDE during two periods (1989-91, 1991-93), the AR during two periods (1999-2000, 2000-01), the JN during one period (2005-07), the PF during one period (2005-07), and the PRO during one period (2005-07).
This restricted analysis underscores the tendency of the governing party to govern/vote alone. Only during the 1999-2000 period did another party (with at least 3 percent of the seats) accompany the governing party in the Government Bloc, with FREPASO joining its Alianza partner, the UCR.
This analysis also highlights the general absence of relevant parties in the Center Bloc, with no party in this zone in eight of the 12 periods. In only four instances we do find a party in this category. In 1993-95 the UCEDE occupied this space while in 2000-01 FREPASO was located in the Center Bloc, as was the AR which moved into this category from the Opposition Bloc when Domingo Cavallo became President De la Rúa's Minister of Economy. Finally, in 2005-07 the Peronist satellite party, the PF, had a voting record in the Chamber which placed it firmly in the Center Bloc.
This same analysis underscores the tendency of most non-governing parties to occupy a position in the Opposition Bloc throughout this period. Of the ten periods where at least one party other than the PJ or UCR held at least 3 percent of the seats, in five an absolute majority of the parties were in the Opposition Bloc. In two others (2001-03, 2003-05), an absolute majority of the parties were in either the Opposition Bloc or the Ultra Opposition Bloc. Two of the three final cases come from the Alianza period, with two of four parties in the Opposition Bloc in 1999-2000 (PJ, AR) and two of five parties in the Opposition Bloc in 2000-01 (PJ, ARI). The last case is 1991-93 period when the UCR was alone in the Opposition Bloc, with the UCEDE in the Center Bloc, and the PJ in the Government Bloc. Lastly, the Ultra Opposition Bloc was occupied in only two instances, by the ARI in 2001-03 and 2003-05, underscoring the rare presence of parties in this category.
The experience of the FREPASO and its role in the Alianza illustrate key aspects of the government-opposition dimension present in roll call voting in the Argentine Congress. The FREPASO was a relevant actor in the Chamber of Deputies between 1995 and 2001, occupying between 9 percent (1995-97) and 15 percent (1997-99) of the seats (see Table 1). During this period, the FREPASO formed an electoral (but not legislative) alliance with the UCR for the 1997-99 period, and finally established an electoral, legislative, and governmental alliance with the UCR for most of the 1999-2001 period, during which time the UCR-FREPASO Alianza was the majority “party” in the Chamber as well as occupied the presidency (i.e., President De la Rúa of the UCR). During the 1995-99 period, when the UCR and FREPASO were both minority parties in the Chamber and were not the party of government (which was President Menem's PJ), the UCR and FREPASO had very similar median ideal points (see Figure 1).
In 1999 the Alianza took control of the presidency and of the Chamber. As such it faced many of the dilemmas that the PJ had faced during the previous ten years: the need to pass legislation that was not always popular with the public or with many party members. Confronted with a UCR President who often did not consult with the FREPASO regarding major policy decisions, and more comfortable playing the much easier role of the opposition in Argentina's adversarial legislative environment, following the resignation of FREPASO leader Vice President Carlos “Chacho” Álvarez in October 2000, a majority of the FREPASO deputies adopted voting positions that placed the delegation significantly more towards the opposition end of the government-opposition dimension than the UCR delegation in the 2000-01 period. Despite growing unrest within the UCR over the direction of the De la Rúa Administration, the UCR deputies remained relatively homogenous (see Figure 5) as well as generally supportive (at least on those issues that made it to the floor for a roll call vote) of the De la Rúa Administration (up until its final two months in office).
Intra-Alianza Heterogeneity, 1995-2001
The Alianza in Opposition, 1997-1999
Figure 4 concisely chronicles the history of the UCR-FREPASO alliance. In 1995-97 (when no alliance was in place), and 1997-99 (when only a partial electoral alliance had been in force for the 1997 elections, but not for interparty relations in the Chamber), the ideal point (and 95 percent confidence interval) of virtually all of the FREPASO deputies were encompassed by the 95 percent confidence interval of the median UCR deputy. While outliers in the FREPASO (and the UCR) increased during the 1999-2000 period, 92 percent of the FREPASO deputies (and 94 percent of UCR deputies) were still encompassed by the median UCR deputy. In the 14 months following Vice President Álvarez's resignation (i.e., the 2000-01 period) however, a large proportion of the FREPASO deputies had ideal points (and 95 percent confidence intervals) that were no longer encompassed by the median UCR deputy. While the percentage of UCR deputies encompassed by the median UCR deputy remained steady at 94 percent, the comparable figure for the FREPASO deputies dropped from 92 percent to 32 percent.
In sum, the Alianza worked very well when both political parties were in the opposition and united by their hostility toward the PJ government of President Menem. When however this UCR-FREPASO Alianza became responsible for actually governing (through its control of the presidency and of the Chamber), it showed itself to not be up to the task (Llanos and Margheritis 2006). While the UCR remained united (albeit increasingly grudgingly so in 2001), less than a third of FREPASO deputies maintained ideal points close to the UCR median. The result was a median ideal point gap for the two parties of 0.61 (UCR: −0.73, FREPASO: −0.12), when the difference during the 1999-2000 and 1997-99 periods had been 0.14 and 0.02, respectively. The median ideal point differences between the 1997-99 and 2000-01 periods are especially noteworthy, since in contrast to the 1999-2000 and 2000-01 periods, the UCR and FREPASO maintained completely separate delegations in the Chamber (with no institutionalized mechanisms of joint-consultation) during the 1997-99 period. 21
During the 1999-2000 and 2000-01 periods, FREPASO's Darío Alessandro was the Alianza Chamber delegation leader (as well as the FREPASO sub-delegation leader).
The impact of having to assume the responsibilities of government on the Alianza is displayed in graphic detail in Figures 5, 6, and 7. Figure 5 shows the distribution of the Alianza deputies’ median ideal points (UCR: circles, FREPASO: triangles) and 95 percent posterior confidence intervals (UCR: uncapped lines, FREPASO: capped lines) during the 1997-99 period, while Figures 6 and 7 show the same distribution for the 1999-2000 and 2000-01 (i.e., after Álvarez's resignation) periods, respectively. Whereas the deputies from the two parties were relatively interspersed during the 1997-99 and 1999-2000 periods, in the 2000-01 period notable separation occurred. In 2000-01, the FREPASO dominated the positive end of the Alianza distribution and the UCR the negative end as well as the middle, underscoring the slow death experienced by the Alianza during its final year of existence. The defeat suffered by the Alianza in the 2001 congressional elections resulted in its loss of majority status in the Chamber on December 10, and ten days later the governmental option represented by the Alianza ended in abject failure with President De la Rúa's resignation.
The Alianza in Government I: The Salad Days, 1999-2000
The Alianza in Government II: The End of Days, 2000-2001
Figure 8 provides the median deputy ideal point and 95 percent confidence interval for the nine largest legislative delegations in the Chamber of Deputies during the 2005-07 legislative period. 22 One extreme of the dimension is occupied by the governing PJ with a median ideal point of 0.53. No other party comes close to having its 95 percent confidence interval overlap with the PJ's 95 percent confidence interval.
The number of delegation members who voted on more than 20 percent of the roll call votes held are included in Figure 8 in parentheses next to the party acronym. The total number of members who voted on over 20 percent of the votes is 265, reflecting the replacement of some sitting deputies by alternate deputies as the tenure of the legislative period progressed.
Median Party Legislative Delegation Ideal Points, 2005-2007
Moving from left to right on this government-opposition dimension, the party most closely adjacent to the PJ is the PF. While the PF was considered to be a satellite party of the PJ during this period, and in fact at the start of the 2007-09 legislative period its remaining members joined the official PJ bloc, the PF's median ideal point (and 95 percent confidence interval) indicates that during the 2005-07 period its membership voted in a manner that was distinct from that of the PJ deputies in aggregate. All the same, however, the PF's location on this dimension is quite different from that of all of the other parties, especially those of the Opposition Bloc. Only the also pro-Kirchner (until late 2007) Partido Nuevo Contra la Corrupción (PNCC) from the province of Córdoba possessed a 95 percent confidence interval that overlapped with that of the PF. The final Center Bloc party is the Partido Socialista (PS), four of whose deputies came from the province of Santa Fe.
The remaining parties were all firmly in the Opposition Bloc. The core of the Opposition Bloc is occupied by the ARI, UCR, and PRO, who have essentially identical median ideal points. The extremes of the Opposition Bloc are represented by the MPN on the government side of the dimension with a median ideal point of −0.33 and the main Peronist opposition to President Kirchner, the JN, on the opposition side at −0.59.
Another perspective from which to view this dimension is by the relationship of the parties to the Kirchner Administration in terms of the 2007 presidential and provincial elections. The PF's members supported the presidential candidacy of the PJ's Cristina Fernández de Kirchner and, returning to the Peronist-fold, occupied positions on the PJ (Frente para la Victoria) electoral lists at the national and provincial levels.
The leaders of the PNCC (Luis Juez) and PS (Hermes Binner) adopted a non-confrontational position vis-à-vis the Kirchner Administration during most of the 2005-07 period. This was part of their respective broader strategy designed to curry favor with President Kirchner in an attempt to either gain his support for their gubernatorial candidacy, or at least obtain his neutrality or lukewarm support for the principal rival in their respective provinces of Córdoba and Santa Fe (who in each case was from the PJ). 23
Juez broke with Kirchner after losing the Córdoba gubernatorial election on September 2, 2007 because Kirchner did not support Juez's challenge of his defeat. While Binner (who was elected governor of Santa Fe on September 2) adopted a neutral position in the October 28 presidential contest, PS Senator Rubén Giustiniani was Carrió's vice presidential candidate.
All of the remaining parties represented the strongest opposition to the PJ in the 2007 national and/or provincial elections. Most supported ardent Kirchner/Fernández de Kirchner opponents in the presidential election (i.e., Elisa Carrió, Roberto Lavagna, Alberto Rodríguez Saá). For instance, the UCR aligned behind the presidential candidacy of Lavagna (as did some members of the JN; others backed the candidacy of PJ dissident Rodríguez Saá) and the ARI supported the candidacy of its founder (Carrió). The PRO did not participate in the presidential election, but in 2007 fielded anti-Kirchner candidates at the provincial level, with its leader, Mauricio Macri, defeating the Kirchner-backed candidate (Daniel Filmus of the PJ) in the June 24 Capital Federal chief of government/mayoral runoff election.
Employing a unique source of information on partisan interaction, this study has presented empirical evidence of the nature of inter-party dynamics in the Argentine Chamber of Deputies between 1989 and 2007. These data provide several important insights on the Argentine legislative party system.
During the past twenty years, roll call voting in the Argentine Chamber of Deputies has occurred primarily along a single policy dimension that discriminates among the parties based on their relative position vis-à-vis the legislative agenda of the governing party. Under this interpretation, the floor votes separate deputies into government and opposition camps (of varying intensity) based on their partisan affiliation, with the governing party (also the majority in the Chamber) functioning as a cartel that effectively controls the agenda in both a negative and positive manner, reducing the ability of the opposition parties to influence policy. Furthermore, given the fact that much of the legislation being discussed during this time period was unpopular with voters, a common response by the non-governing parties has been to assume a consistent opposition to the majority party's legislative agenda, given the reputational benefits that accrue to the party from its challenge of unpopular policies.
When it came to many of the most salient public policy/legislative initiatives of the day (Jones and Hwang 2005), on average the governing party in Argentina has tended to vote/govern alone, with the president's party's (which was always the majority party) median ideal point significantly distinct from those of virtually all other parties. Even Argentina's only experience with a coalition government (1999-2001) revealed the system's inherent majoritarian tendency. Most of the members of the UCR-FREPASO Alianza, except those from the president's party (UCR), had, by the second year of the Alianza's short life span (i.e., 2000-01), begun to vote frequently against the bills promoted by UCR President Fernando de la Rúa and his party's leadership in the Chamber.
Finally, this government-opposition dimension continued in force during the administration of President Néstor Kirchner, with the PJ tending to vote/govern alone. The principal non-governing parties, in spite of their ideological differences, occupied similar positions (on the opposition side of the dimension) on this policy space. For instance, the center right PRO, centrist UCR, and center left ARI had median ideal points during the 2005-07 legislative period that were virtually identical. In a similar vein, during this period, the Center Bloc was occupied by ideologically dissimilar parties such as the center-right PRS, center-left PS, center-left PNCC, and Peronist PF. What united these latter four ideologically distinct parties was however their ties to the Kirchner Administration, with each party's legislative voting behavior strategically designed to obtain the support/not incur the wrath of President Néstor Kirchner.
Roll call vote analysis clearly has its limitations, and a complete understanding of the functioning of the Argentine Congress and legislative party system requires a multifaceted approach that utilizes a variety of different data sources and methods to study these complex institutions (e.g., Alcántara Sáez 2008; Alemán, Calvo, Jones, and Kaplan 2009; Baron 2008; Benton 2002; Calvo 2006; Eaton 2002; Föhrig 2007; Jones 2008; Jones, Saiegh, Spiller, and Tommasi 2002; Leiras 2007; Llanos 2002; Malamud and De Luca 2005; Micozzi 2008; Mustapic 2002; Oliveros and Scherlis 2006; Spiller and Tommasi 2007). In this article we make an important addition to this growing literature, utilizing roll call vote data to provide a better understanding of inter-party dynamics in the Argentine Chamber of Deputies over the past twenty years.