Uprooted but Stable:Chilean Parties and the Concept of Party System In - PDF document

Uprooted but Stable:Chilean Parties and the Concept of Party System Institutionalizationuan Pablo Luna David Altman Mainwaring and ScullyÕs concept of party system institutionalizationexploring the recent evolution of the conceptÕs four dimensions ininstitutionalized (as conventionally argued) but is simultaneouslysociety. In this regard, it approaches some recent descriptions of theroots in society. This article argues that the current operationaliza-ainwaring and ScullyÕs concept of party system institutionalization,first published in 1995, has greatly influenced the literature on par-ties and party systems, to the point where it has become the establishedally, Mainwaring and Scully (1995) identify four dimensions of partya.the stability and regularity of party competition patterns b.the presence of party roots in society, which helps to create thec.the relative level of party legitimacy in societyd.the presence of well-developed party organizations, as opposed tobetween the four dimensions of PSI as positive and linear, arguing that © 2011 University of Miami based on empirical indicators for each of the four dimensions. Alterna-ures of electoral volatility.challenged nor empirically tested (though for significant exceptions seeCrisp 2000; Kitschelt 2000; Lindberg 2007).definition and measurement of PSI (see Torcal forthcoming). Forinstance, Zucco (forthcoming) shows that in the Brazilian case, decreas-system as one that defies a one-dimensional understanding of PSI. party system concept. On the other hand, current empirical indicators ofThis article contributes to the ÒrevisionistÓ PSI literature by explor-on partial and more imperfect operationalizations, which suffer from adiachronic trends and cross-sectional variance regarding different indica-tors of party system institutionalization (which are usually measured at the TIN AMERICAN POLITICS AND SOCIETY53: 2 adequacy of a one-dimensional conceptualization of PSI. At themoment, there is a strong consensus that Chile is an example of a highlytake issue with this portrayal and show that electoral stability(measured as low electoral volatility at the national level) coexists withother characteristics that are typical of noninstitutionalized party sys-tems. In other words, we find that the Chilean party system does notsystem combines very low national-level volatility with low levels ofrootedness and legitimacy in society, and weak partisan organizations.On this basis we argue that a situation similar to that described by Zuccoconfiguration or evolution of PSIÕs dimensions. The Chilean case, how-ever, being the epitome of a highly institutionalized party system, con-of PSI across conceptual dimensions that are theoretically conceived asFor the analysis of PSIÕs four dimensions in contemporary Chile, wefocus first on electoral volatility (as a proxy for the stability of electoralcompetition patterns) at both the national and subnational levels, con-political coalitions and parties (see AgŸero et al. 1998). In addition, wepresent evidence of the recent transformation of partisan organiza-tions in posttransitional Chile, arguing that increasingly personalizedand localized leadership styles are emerging while political parties, asinstitutions, have lost strength. Last, we analyze the legitimacy of polit-ical parties and electoral processes in Chile. The concluding sectionexamines the theoretical implications of our empirical analysis. OLATILITYbility, and is one of the crucial components of any index ofPSI(Main-is the sum of the absolute changes in vote shares divided by 2 (Peder- LUNA AND ALTMAN:CHILEAN PARTIES 3 with patterned interactions of party competition. In this regard, Chilecratic world as a whole, given its low levels of electoral volatility (seemeasure is highly problematical, for at least two reasons. First (andof volatility has problems of internal validity; it might not measurevolatility index systematically commits a gross ecological inference,the number of votes for a party at time T and the number of votesine a perfect bipartisan system with two identical parties, A and B, eachwith 50 percent of the votes. It might occur that about half of the voters TIN AMERICAN POLITICS AND SOCIETY53: 2 able 1. Electoral Volatility in Latin America olatilityPresidentialAverage Lower ChamberVolatilityVolatilityHonduras7.676.236.95Nicaragua15.0510.5012.77Chile3.4722.1712.82Uruguay14.6514.5914.62Costa Rica46.9612.9514.95Mexico14.2318.4316.33El Salvador18.0919.4818.79Dominican Republic19.0923.0821.09Paraguay19.8628.7024.28Argentina18.3531.7025.02Colombia17.5133.6425.58Panama19.3634.7827.07Brazil28.6736.3532.51enezuela28.9837.0433.01Bolivia29.0938.6833.88Ecuador32.5546.2639.41Guatemala46.9548.9547.95Peru51.8352.2152.02verage22.3528.6525.50 Source: Payne et al. of A shift their preferences to B in the next election and otherwise. Inthis case, volatility would be zero, even though 50 percent of the citi-most, PedersenÕs volatility index provides only a lower boundary on thecases for elucidating the evident tension that emerges when selectinginvolves, in this case, choosing as the unit of analysis either individualThat risk derives from the enormous influence the binominal elec-toral system has on the distribution of candidates in each municipalityPPD in election E and a member of the PDC in the next election E+1.(another party included in the center-left Concertaci—n electoral pact)actually might be due to a mirage produced by aggregation bias whencomputing volatility for electoral pacts. Computing by pacts, however,up candidates in every district. In short, both available solutions areindexes at the partisan level, considering different geographical levelsmunicipalities). This takes advantage of the system, in which Chileansticket, Chilean voters can perfectly ÒcutÓ the vote. Bracketing out theinternal validity problems attributed to any computation of the Pedersendifferent levels of analysisÑboth partisan and geographicalÑa com- LUNA AND ALTMAN:CHILEAN PARTIES 5 pletely different picture from that embraced by the conventionalThere is a rather strong consensus that posits that Chile has very lowelectoral volatility (to mention a few: Angell 2003; Mainwaring andScully 1995; Roberts and Wibbels 1999). Yet, as already mentioned, thismeasurement is problematic for our case (compare Morgenstern andPotthoff 2005, 36). Therefore, we compute electoral volatility using dif-ferent units of analysis (parties and coalitions) and different levels ofaggregation (18 senate circumscriptions, 60 lower-chamber districts, andmates suggests a completely different take on ChileÕs electoral stabilitydidates at this level in every municipality) is higher as well. In short,there are some grounds to state that the low volatility obtained at thenational level, which informs comparative assessments of PSI, is proba- TIN AMERICAN POLITICS AND SOCIETY53: 2 Figure 1. Volatility in Chile by Parties and Coalitions Source: AuthorsÕ construction based on electoral results. OOTSINnent in operationalizing this dimension in recent works, such as Main-waring and Torcal 2005 (18).have sought to describe the emerging patterns of partisan competition. Instructure, hypothesizes that after the democratic breakdown, the socialcleavages that were fundamental to the party system would soon reappear,replaced by different scholars (Tironi and AgŸero 1999; Torcal and Main-waring 2003), whose analyses described an emerging cleavage structureIn their highly influential 2003 article, Torcal and Mainwaring arguewere able to recraft the social bases of party competition. They couldaccomplish this, the argument goes, because the class cleavage thatsystem (AgŸero et al. 1998; Carey 2002; Tironi and AgŸero 1999). Torcaland Mainwaring (2003), however, ground their hypotheses in an ambi-and TorcalÕs argument. Table 2 presents different possible proxies forthe programmatic structuring of partisan electorates. These alternativercalÕs 2005 measurement strategy. The first model tries to approximateissues that together represent salient competitive divides: state-market, LUNA AND ALTMAN:CHILEAN PARTIES 7 TIN AMERICAN POLITICS AND SOCIETY53: 2 able 2. Programmatic Structuring of Partisan Electorates (selected countries) SubstantiveSymbolicTotalWavesNo.SubstantiveSymbolicTotalWavesNo. 1990s1990s1990sAvailableParties2000s2000s2000sAvailablePartiesPeru.023.009.0161994Ð993.012.007Both3/3Brazil.018.044.031Both4/6.015.0072005Ð83Colombia.015.025.021994Ð993.016.0072005Ð85enezuela.011.064.0371994Ð993.0008.0272000Ð20042Argentina.027.03.2851994Ð992.022.013Both3/4Mexico.015.038.027Both5/3.013.025Both3/3Chile.046.118.082Both3/5.043.159Both6/6Uruguay.056.25.1531994Ð994.037.3 In Uruguay, the World Values Survey was not applied. The reported results are based on data provided by NŽstor Da Costa, who reWVS questionnaire and sample design in 2002. We thank him for providing access to this dataset. preferences in systems where programmatic linkages are stronger.able in the World Values Survey. This holds true for each of the threepossible operationalizations used. Moreover, if, instead of predictingtwo major coalitions (the Concertaci—n and the Alianza), the indexan R-squared of .38 (which more than doubles the already high coeffi-While at first glance it is consistent with Mainwaring and TorcalÕs2003 assessment, our evidence supports a different view of the Chileanparty system. It suggests a series of theoretical qualifications to Torcaland MainwaringÕs 2003 argument about the top-down creation of polit-ical cleavages in transitional settings. At least when applied to Chile, weMethodologically, Torcal and Mainwaring anchor their analysis ofing to LAPOP 2010 results), party identifiers are not representative of thethe current dynamics of the party system. We need to incorporate aThird, Torcal and Mainwaring took political parties as the relevantinto account political pacts (the Concertaci—n and the Alianza) as units.While Torcal and MainwaringÕs specification does justice to the histori-scenario. Taking political pacts as units, we also lower the bar for find- LUNA AND ALTMAN:CHILEAN PARTIES 9 Substantively, our re-analysis of the Chilean party system suggests aseries of qualifications to the characterization of the posttransitionalsystem as ideologically rooted. The authoritarian-democratic divide istime (which is consistent with Torcal and MainwaringÕs expectation) andtends to mobilize specific segments of the Chilean electorate more effec-mobilization. Moreover, voters who are less politically socialized (partic-ing of political competition by political parties. Moreover, the allegedMost Chilean citizens do not sympathize with any political party,2010). Furthermore, according to recent survey evidence, the percent-age of Chileans declaring to sympathize with a party declined from 252006Ð10). This situation calls into question Torcal and MainwaringÕsAlthough the levels of partisan disengagement are high across allsocial segments, interesting variance is observed when segmenting thecitizens are homogeneously disengaged, middle-educated citizens, withtisan sympathy. The older cohort in the highly educated segment is theone that presents a greater proportion of party sympathizers, surpassing30 percent. The youngest cohort in that education tercile also presentsrelatively high levels of partisan sympathy. However, middle-aged andIn sum, low- and medium-educated citizens and youth (with the excep-of partisan attachment.responses to a follow-up question on their voting behavior, discriminat-ing between those who do not have a party but still vote and those who 10LATIN AMERICAN POLITICS AND SOCIETY53: 2 Results are once again shown for different age cohorts and for thehigher (HE) and lower (LE) education terciles. Whereas those who doyounger age cohorts. This evidence is consistent with declining voterturnout (and voter registration) among the youth (see Carlin 2006; ToroMaureira 2008). However, partisan disengagement is also very highIn spite of all the trends we have analyzed so far, comparative evi-vote choice in Chile. Even in a context of weak and declining partisanable 3 presents the correlation between the left-right scale andcorrespond to the third and fourth waves of the World Values Survey LUNA AND ALTMAN:CHILEAN PARTIES 11 Figure 2. Sympathy with Political Parties reports specific correlation coefficients for the most and least educatedsegments of the population and computes the gap between them.of the time periods for which data are available, we would rather insistauthoritarian period and the transition to democracy continues to erode. The moral issue fluctuates in terms of both the strength of the cor-relation and the size of the gap between different educational cohorts,but it never maps strongly on left-right self-placements. Furthermore,the state-market divide presents stabler but marginal correlations withthe left-right scale (especially for low-educated cohorts). It also presentssistently than their less educated fellow citizens. It therefore seems that 12LATIN AMERICAN POLITICS AND SOCIETY53: 2 Figure 3. Reported Electoral Behavior and Relationship to Parties by Age Cohorts and Education Levels LE = lower education terciles; HE = higher education terciles. complement this analysis, we now turn to citizensÕ own expla-nation of their voting decision at the district level (see table 4). Thosetraits or congressional campaign platform) for their vote. Meanwhile,This evidence calls into question current assessments of the strengthof partisan identities in the system. It also challenges estimations of pro-sympathize with a political party. While partisan dealignment is usuallyof party systems with strong roots in society. Indeed, compared with thevidual party level might be too pessimistic, given the strong role LUNA AND ALTMAN:CHILEAN PARTIES 13 able 3. Correlations Between Left-Right Self-Placements RegimeMoralState-Market (+Left=+Democratic)(+Left=+Liberal)(+Left=+Statist)Overall Correlation.62.17.08Highly Educated R.75.20.42Social Gap (LE-HE)Ð.12.02Ð.33Overall Correlation.28.06.11Highly Educated R.18.15.27Social Gap (LE-HE)Ð.04Ð.26Ð.19Overall Correlation.27.13.11Highly Educated R.26.17.18Social Gap (LE-HE)Ð.02Ð.03Ð.11Overall Correlation.21.07.09Highly Educated R.37.15.25Social Gap (LE-HE)Ð.22Ð.09Ð.15 Source: AuthorsÕ construction based on Luna 2006; LAPOP. Alianza (Alliance) is composed of two right and center-right parties, thethe national government in 1990 and held it until 2010. It is composedof four parties: Partido Dem—crata Cristiano (DC, Christian Democrattransition to democracy. To test this alternative explanation of dealign-citizens declaring adherence to none of those political coalitions haveAre these trends toward ideological destructuring innocuous in termsof citizensÕ symbolic attachment to different ideological or programmatic 14LATIN AMERICAN POLITICS AND SOCIETY53: 2 able 4. Explanations of Vote Choice Education Level_____________________________________ PrimarySecondaryUniversityTotalThe candidate (individual traitsand campaign platform)94.791.483.390.9candidate5.08.316.28.8Other reasons0.30.20.50.3otal100.0100.0100.0100.0 Question: Which of the following was the most important factor in your vote? Source: AuthorsÕ construction based on LAPOP 2006. currents included in the CEP survey (left, center-left, center, center right,Thus, although they display comparatively high levels of program-dence elaborated in this section is also consistent with qualitative nar-ratives on partiesÕ increasing split from grassroots organizations andinterest groups (Oxhorn 1995; Posner 2004). RGANIZATIONSComparative indicators of party organizational strength are not easilyorganizational penetration of society, we empirically tackle this chal-lenge by first drawing on qualitative evidence drawn from previous LUNA AND ALTMAN:CHILEAN PARTIES 15 Figure 4. Percentage of Identification with the Two Main Coalitions, Notes: The scores for the year 1991 are calculated by aggregating partiesÕ scores. The 2002; Garret—n 1988; Valenzuela 1977). It has been argued that politicallocal politics following the return to democracy by centralizing candi-However, the evidence we present in this section suggests that thenature of those networks has been transformed in the posttransitionalnomination prerogatives have allowed partisan elites to remain power-inations strategically to maximize the electoral return of their allocationquotas, remaining powerful by rewarding Òparty loyalistsÓ by prioritiz-ing their candidacies in interparty negotiations. However, the positionsof incumbent candidates who want to seek re-election are virtuallyuncontested (Carey and Siavelis 2005; Navia 2008).Indeed, incumbentelections), and their electoral fortunes hinge more on their performanceties (Navia 2008). Moreover, removing successful incumbents, even ifConcomitantly, party organizations (and the value of party labels)omy of mayors provided by municipal decentralization and statereforms has also limited the capacity of partisan organizations to cen-tralize power by maintaining in place hierarchical networks tying local,cific nuances, converging leadership characteristics at the municipal,and increasingly need to set up parallel organizations to serve their dis-nature, but are fundamentally centered on the personality of the con-gressional member. This explains the levels of subnational electoralvolatily we showed in the first empirical section of this article. Indeed,run for a senate seat, I do not think those votes are transferable toanother PS candidate, even though I have people working with me.The same occurs with candidates of the right. 16LATIN AMERICAN POLITICS AND SOCIETY53: 2 The moment she quit RN to join UDI, RN disappeared in the dis-trict at the congressional level. We keep the vote at the municipalsupport base. . . . The vote here is personal; no one is voting forening of partisan organizations. In 2005, for the first time since 1990,the independent senator Carlos Bianchi, from region 12, was elected.Moreover, since 2006, four senators from the Concertaci—n have(including former DC president Adolfo Z‡ldivar) defected explicitly,held since 1990. Several deputies have also renounced their party affil-iation or have made public their differences with the coalition. Thecarrying a formal party nomination also contributes to this emergingcohesive party organizations. never been. People who used to throw stones at me and who sentThe emerging configuration at the district and local levels alsoreduces partiesÕ capacity to structure consistent programmatic platformsThey donÕt take issue with anything. You can see Congress mem-bers taking issues, but parties, as political parties, never. You donÕttogether to defend their individual interests. (Toh‡ 2003)At the municipality level, the pro-independent trend is stronger. Theernment has surely contributed to this trend. Once the electoral system LUNA AND ALTMAN:CHILEAN PARTIES 17 applied at the municipal level stopped penalizing independent candi-efited from a leadership renewal promise and from a campaign com-new generation. This promise was made explicit by the slogan, ÒIn mygovernment, none of those who have been in office so far will partici-Marco Enr’quez-Ominami, splitting from the Socialist Party after theparty president neglected his request to take part in the Concertaci—nthe vote in the December 2009 presidential election. During the cam-paign, this unexpected candidacy posed a significant threat to the can-didacies of former president and traditional Concertaci—n leaderEduardo Frei and the Alianza challenger, Sebasti‡n Pi–era. ized, giving rise to the emergence of independent candidates, whosometimes compete on an antiÐparty politics platform. This results from 18LATIN AMERICAN POLITICS AND SOCIETY53: 2 Figure 5. Levels of Citizen Participation in Electoral Campaigns Question: Did you work for any candidate or party in the last presidential elections? and contributes to the weakening of partisan organizations as institu-(or at least the social penetration) of partisan organizations in Latinthe arguments presented in this section. As figure 5 shows, in the LatinAs Mainwaring and Scully (1995) argue, we should expect parties andelectoral processes to display relatively high levels of legitimacy in thecontext of an institutionalized party system. This last empirical sectionracy can function without political parties.Ó The case is placed among LUNA AND ALTMAN:CHILEAN PARTIES 19 Figure 6. Need for Political Parties Question: Democracy can exist without political parties (1 = strongly disagree Ð 7 = Brazil tion of responses obtained in different countries (all of which haveseen recent and significant partisan turmoil) when citizens were askedabout different ways of seeking changes in the countries they lived in.political changes to voting behavior, and the greatest percentage ofresponses to public opinion surveys. Actual electoral behavior, however,seems consistent with this evidence. Figure 7 presents data on electoralparticipation from the beginning of the transition in 1988 to 2008. Thisfigure bluntly expresses the decay in the most basic form of civicpercent), this is caused by the sole consideration of registered voters(Mainwaring and Scully 2010). However, registration in Chile is an indi-for those who are registered. Once we calculate different measures for7 shows, in different possible operationalizations, electoral participationThus, the evidence in this section lends preliminary support to aprocesses, once again contradicting the expectation for highly institu- 20LATIN AMERICAN POLITICS AND SOCIETY53: 2 able 5. Ways to Seek Changes in a Country ColombiaEcuadorBoliviaPeruVenezuelaChiledefend your interests433850437645movements1421131898other means152519241016changing things28151815531 Source: LAPOP 2008. Let us first issue a warning against some possible misinterpretations ofour argument in the case. Conventional wisdom on policymaking inChile (Stein et al. 2006; Mainwaring and Scully 2010; World Bank 2010)ence of a highly institutionalized party system.do not argue againstsuch portrayals of the policymaking process; instead, we claim that itsness, legitimacy, and party organizational features). Second, though wetutionalization in most of the alternative indicators to electoral volatilitythat we have discussed here, we do not necessarily contend that Chile LUNA AND ALTMAN:CHILEAN PARTIES 21 Figure 7. Electoral Participation in Chile, 1988Ð2008 Note: VAP means voting-age population. matter, normatively better) than the current system. are more interested in the analytical implications of the Chileanteleological or normative speculations on the historical and future evo-Indeed, we have shown that in Chile, decreasing national-level electoralvolatility combines with decreasing party roots in society, decreasinglegitimacy, and the weakening of partisan organizations as institutions. In our view, the nonlinear relationship between PSI indicators wehave described can be interpreted in at least two ways. Those alterna-tive readings have different implications for the concept and this case.However, we will argue that both readings are possibly complementary.dimensional (positive linear) conceptual structure proposed by Main-waring and Scully is not challenged. Instead, the inconsistency relates toFor instance, comparing the evolution of electoral volatility over time aswhile ignoring that increased (Brazil) or constant (Chile) stability mightcases. Venezuela in the 1990s, a case described as institutionalized infew years down the road (Morgan 2007). In sum, both the temporal evo-lution of electoral volatility in regard to other dimensions of PSI in casessuch as Brazil, Venezuela, and Chile, and the evidence presented forindicator for other dimensions of PSI, and is, by far, the one most fre- 22LATIN AMERICAN POLITICS AND SOCIETY53: 2 as in Chile, in a reasonable number of additional empirical instances, itmight be necessary to consider the substitution of electoral volatility assented in relation to the concept of PSI is to argue that the conceptualseveral dimensions that might relate to each other in nonlinear ways. Inthings go togetherÓ assumption, and thus neglects potential trade-offsbetween different dimensions of the concept. Electorally stable but oth-the concept. Intertemporal trade-offs between dimensions are alsoobscured. For instance, if we take volatility as a leading indicator forthat pursue a successful realignment (one that temporarily creates someor legitimacy) are penalized. At the same time, ÒfrozenÓ party systems,which are stable but become increasingly detached from civil society, getsufficientÓ dimensions (see Goertz 2005). Although it is more intensiveoretically entertain the possibility of observing (perhaps for extendedimmediate (or inescapable) collapse. Moreover, a multidimensional def-inition of PSI could also integrate, as dimensions, emerging concepts inThe case of Chile illustrates the different implications that could bederived from these two alternative readings regarding concept structurequently described and admired (Mainwaring and Scully 2010). However, LUNA AND ALTMAN:CHILEAN PARTIES 23 the case of Chile would approximate, in a two-dimensional conceptualstructure, the ideal type of stable but uprooted party systems (a), lackingone of the necessary and sufficient conditions (rootedness) to be con-sidered fully institutionalized. Such configuration differs from three otherconform to the Òall good things go togetherÓ assumption (and teleology)built into the original concept, types a and d might eventually be neededconfiguration and evolution of contemporary party systems. cant improvements. If confirmed by future research, the presence of aconceptual and measurement refinements, building on Mainwaring and1.See Luna forthcoming for an extensive argument along these lines.2.For a discussion on the concept of Òecological fallacy,Ó see King 1997.3.The tension on the units of analysis to study has been a constant4.This and the following section draw on evidence presented in Luna5.ÒTheoretically, we prefer the information about party identificationbecause party realignment should be measured by stable elements of party pref-on short-term changes in public opinionÓ (Torcal and Mainwaring 2003, 60).6.The questions representing these three issues are the following: a) 1Ð10with the following statement: ÒWe need larger income differences as incentivesfor individual effort.Ó b) 1Ð4 scale reflecting agreement with the following state-ment: ÒDemocracy may have problems but it is better than any other form ofgovernment.Ó c) 1Ð10 scale with 1 representing ÒAlways justifiableÓ and 10umentation of this analysis is available from the authors.7.A similar result is presented by Ruiz-Rodr’guez (2003), who shows that8.While the 1995 time frame might be consistent with Torcal and Main-waringÕs claim that political cleavages might be transitory, from a conceptualstandpoint, this is virtually a contradiction in terms, because cleavages are usu- 24LATIN AMERICAN POLITICS AND SOCIETY53: 2 waring and TorcalÕs 2005 notion of Òpolitical cleavageÓ for the less restrictivetent with the transitory nature of posttransitional alignments that Torcal andoverhauled its cleavage structure.9.Full documentation on these factor analyses is available in Luna 2006. 10.During her electoral campaign, Bachelet announced, Ò espero que enmi gobierno ningœn ministro se repita el plato.Óseminar organized by the Asociaci—n Nacional de Empleados Fiscales (ANEF),El Mercurio11.This blatant decay persists even without considering the plebiscite of1988 or the elections of 1989. Indeed, as the figure illustrates, the coefficient ofdetermination, R2, for the complete series is 0.75; without considering the afore-mentioned votes, it remains highly statistically significant at 0.64. 12.See, e.g., World Bank 2010.AgŸero, Felipe, Eugenio Tironi, Eduardo Valenzuela, and Guillermo Sunkel.1998. Votantes, partidos e informaci—n pol’tica: la fr‡gil intermediaci—nCarey, John M. 2002. Parties, Coalitions, and the Chilean Congress in the 1990s., ed. Scott Morgenstern and BenitoNacif. Cambridge: Cambridge University Press. 222Ð53.Cavedo, JosŽ Antonio. 2003. Former mayor and current council member, SanCentro de Estudios Pœblicos (CEP). Encuesta permanente, 1991Ð2008.&#xwww.;Îpc;&#xhile;&#x.cl/; ms/;&#xlang;&#x_1/e;&#xncue;&#xstas;ÎP1;.7;.htm LUNA AND ALTMAN:CHILEAN PARTIES 25 Crewe, Ivor. 1983. The Electorate: Partisan Dealignment Ten Years On. European PoliticsCrisp, Brian F. 2000. of Venezuelan Politicians and Interest GroupsDalton, Robert J., Ian McAllister, and Martin Wattenberg. 2000. The Conse-quences of Partisan Dealignment. In , ed. Dalton and Wattenberg.Guerra, Sergio. 2003. Renovaci—n Nacional Council Member, Pe–alolŽn. AuthorInglehart, Ronald. 1984. The Changing Structure of Political Cleavages in West-ern Society. In Beck. Princeton: Princeton University Press. 24Ð69.Jones, Mark P. 2005. The Role of Parties and Party Systems in the Policy MakingProcess. Presented at the conference ÒState Reform, Public Policies, andPolicymaking Processes,Ó Inter-American Development Bank, Washington,A Solution to the Ecological Inference Problem: ReconstructingIndividual Behavior from Aggregate Dataveys. Nashville: V&#xwww.;&#xvand;rbi;&#xlt.e; u/l; pop;anderbilt University. Lindberg, Staffan I. 2007. Institutionalization of Party Systems? Stability and Flu-Government andLuna, Juan Pablo. 2006. Party-Voter Linkages in Two Institutionalized Party Sys-Reforma de los partidos pol’ticos en ChileFontaine, Ignacio Walker, Jorge Navarrete, and Cristi‡n Larroulet. Santiago:necesario un nuevo concepto? In Torcal forthcoming. Mainwaring, Scott. 1999. Rethinking Party Systems in the Third Wave of Democ-Mainwaring, Scott, and Mariano Torcal. 2005. Party System Institutionalizationand Party System Theory After the Third Wave of Democratization. Work-ing Paper 319. Notre Dame: Kellogg Institute for International Studies, Uni-versity of Notre Dame. April. 26LATIN AMERICAN POLITICS AND SOCIETY53: 2 Mainwaring, Scott, and Timothy R. Scully. 1995. Introduction: Party Systems inLatin America. In Democratic Governance inEl Mercurio Morgan, Jana. 2007. Partisanship During the Collapse of the Venezuelan PartyLatin American Research ReviewMorgenstern, Scott, and Richard F. Potthoff. 2005. The Components of Elections:District Heterogeneity, District-Time Effects, and Volatility. , ed. Peter Siavelis and Scott Morgenstern. Univer-Oxhorn, Phillip. 1995. Organizing Civil Society: The Popular Sectors and thePayne, Mark, Daniel Zovatto, Fernando Carrillo, and AndrŽs Allamand. 2003. pol’tica importa: democracia y desarrollo en AmŽrica Latina. Washington,DC: International Institute for Democracy and Electoral Assistance, Inter-Payne, Mark, Daniel Zovatto, Fernando Carrillo, and Mercedes Mateo D’az.La pol’tica importa: democracia y desarrollo en AmŽrica Latinaashington, DC: International Institute for Democracy and Electoral Assis-tance, Inter-American Development Bank. Pedersen, Mogens. 1983. Changing Patterns of Electoral Volatility in Europeanestern Euro-Mair. Beverly Hills: Sage. 29Ð66.Posner, Paul. 1999. Popular Representation and Political Disattisfaction in ChileÕsJournal of Interamerican Studies and World AffairsÐÐÐÐÐ. 2004. Local Democracy and the Transformation of Popular ParticipationRoberts, Kenneth, M., and Erik Wibbels. 1999. Party Systems and Electoralolatility in Latin America: A Test of Economic, Institutional, and StructuralRuiz-Rodr’guez, Leticia. 2003. Polarizaci—n en el Chile postautoritario. Elites par-Pol’tica, dinero e institucionalizaci—n partidista en AmŽrica LUNA AND ALTMAN:CHILEAN PARTIES 27 Rethinking the Center: Cleavages, Critical Junctures,Siavelis, Peter. 2000. The President and Congress in Post-Authoritarian Chile:Stein, Ernesto, Mariano Tommasi, Koldo Echebarr’a, Eduardo Lora, and MarkThe Politics of Policies: Economic and Social Progress in. Washington, DC: Inter-American Development Bank.rcal, Mariano, and Scott Mainwaring. 2003. The Political Recrafting of SocialBritish Journal of Political Sci-ro Maureira, Sergio. 2008. J—venes y generaciones pol’ticas en el Chile post-autoritario. Una mirada a travŽs de los resultados de LAPOP. Political Brokers in Chile: Local Government in a Cen-. Durham: Duke University Press.orld Bank. 2010. Worldwide Governance Indicators. org/gover&#xhttp;&#x://i;&#xnfo.;&#xworl;යn;&#xk.00;nance/wgi/index.asporld Values Survey. Multiple years. &#xhttp;&#x://w;&#xorld;&#xvalu;ssu;&#xrvey;&#x.or1;.80;gZucco, CŽsar. Forthcoming. Estabilidad sin ra’ces: institucionalizaci—n de sis-temas de partidos en Brasil. In Torcal forthcoming. 28LATIN AMERICAN POLITICS AND SOCIETY53: 2 http://onlinelibrary.wiley.com/doi/10.1111/j.1548-2456.2011.00115.x/abstract http://onlinelibrary.wiley.com/doi/10.1111/j.1548-2456.2011.00115.x/abstract