ArticlePDF Available

Status Quo Characteristics or Private Sector Affinity? Explaining Preferences for Safety Nets in Latin American Legislatures

Authors:

Abstract and Figures

The literature on legislative behavior has evidenced a decay of partisanship around the world. In Latin America, where party systems are generally weak, this is a long-standing trend. However, it does not mean that legislators do not group themselves or assume identities to ease their cognitive formation of policy preferences. What shapes legislators’ atti- tudes toward safety net policies in Latin America? We test two hypotheses to predict these cues. First, descriptive representation. We assume that legislators who share a status-quo profile, i.e. older, wealthy, and highly-educated males, tend to share policy preferences. Second, business interests. We test whether having high-level jobs in the private sector and trusting employers’ organizations shape policy preferences. We conduct this investigation based on a dataset with interviews carried out with a large number of legislators from 17 Latin American countries. Our policies of interest are the ones that produce a stark cleavage in the region: The State’s role in the provision of a safety net to the poorer strata of the population. We confirm the first hypothesis and show that descriptive representation has a stronger predictive power of policy preferences than strong ties with the business sector at least when it comes to the selected policies.
Content may be subject to copyright.
STATUS QUO CHARACTERISTICS OR PRIVATE
SECTOR AFFINITY? EXPLAINING PREFERENCES FOR
SAFETY NETS IN LATIN AMERICAN LEGISLATURES1
JOÃO VICTOR GUEDES-NETO
ASBEL BOHIGUES
KEYWORDS
Political Elites; Latin American Politics; Comparative Legislative Studies; Descriptive Representation;
Business Interests.
1 We thank the anonymous referees for their helpful comments, and the editors for their professionalism
organizing this special edition.
ABSTRACT
The literature on legislative behavior has evidenced
a decay of partisanship around the world. In Latin
America, where party systems are generally weak,
this is a long-standing trend. However, it does not
mean that legislators do not group themselves or
assume identities to ease their cognitive formation
of policy preferences. What shapes legislators’ atti-
tudes toward safety net policies in Latin America?
We test two hypotheses to predict these cues. First,
descriptive representation. We assume that legisla-
-
thy, and highly-educated males, tend to share poli-
cy preferences. Second, business interests. We test
whether having high-level jobs in the private sector
and trusting employers’ organizations shape policy
preferences. We conduct this investigation based
on a dataset with interviews carried out with a large
number of legislators from 17 Latin American coun-
tries. Our policies of interest are the ones that pro-
duce a stark cleavage in the region: The State’s role
in the provision of a safety net to the poorer strata of

show that descriptive representation has a stronger
predictive power of policy preferences than strong
ties with the business sector at least when it comes
to the selected policies.
STATUS QUO CHARACTERISTICS OR PRIVATE SECTOR AFFINITY? EXPLAINING
PREFERENCES FOR SAFETY NETS IN LATIN AMERICAN LEGISLATURES
16
CARACTERÍSTICAS DO STATUS QUO OU AFINIDA-
DE COM O SETOR PRIVADO? EXPLICANDO PREFE-
RÊNCIAS POR REDES DE SEGURANÇA SOCIAL NAS
LEGISLATURAS LATINO-AMERICANAS
PALAVRAS-CHAVE
Elites Políticas; Política Latino-Americana; Comportamento Legislativo Comparado; Representação
Descritiva; Interesses Empresariais.
SOBRE OS AUTORES
JOÃO VICTOR GUEDES-NETO
Doutorando da Pittsburgh University – USA. Contato: jog118@pitt.edu
ASBEL BOHIGUES
Doutorando da Universidade de Salamanca-ES. Contato: asbogar@usal.es
SUBMETIDO EM
31/05/2018
APROVADO EM
31/07/2018
RESUMO
A literatura sobre comportamento no legislativo tem
evidenciado a queda do partidarismo ao redor do
mundo. Na América Latina, onde sistemas partidários
geralmente são fracos, essa é uma tendência antiga.

se agrupam ou assumem identidades que facilitam
seus processos cognitivos de formação de preferên-
cias sobre políticas públicas. O que forma as atitudes
de legisladores em relação às políticas de seguridade
social? Nós testamos duas hipóteses para prever estes
atalhos. Primeiro, a representação descritiva. Nós as-

de status quo, em essência homens mais velhos, ricos,
e altamente educados, tendem a compartilhar as mes-
mas prefencias sobre poticas públicas. Segundo, os
interesses empresariais. s testamos se ter empregos
-
zações patronais afetam as preferências por políticas
públicas. Nós realizamos esta investigação com base
em um banco de dados com entrevistas realizadas
com um grande número de legisladores de 17 países
latino-americanos. Nossas políticas de interesse são
aquelas que produzem uma das maiores divisões na
região: o papel do Estado na provisão de uma rede de
seguridade social para as porções mais pobres da po-
-
tramos que a representação descritiva tem um poder
preditivo mais forte sobre preferências sobre políticas
públicas que a forte relação com o setor empresarial
pelo menos no tocante as políticas selecionadas.
17
TEORIAE SOCIEDADE Nº 26.2 - JULHO - DEZEMBRO DE 2018
1. INTRODUCTION
What shapes legislative elites’ attitudes toward safety-net policies in Latin America?
The current literature on political parties has been clear at evidencing a decay on partisanship
(e.g., HOOGHE and STOLLE, 2005; KOVACHEVA, 2005; DALTON, 2008). However, poli-
tical behaviorists constantly point out at the role of identity in predicting attitudes toward a
series of events (e.g., ATKENSON and CARRILLO, 2007). In this sense, it becomes natural to
infer that whereas people are detaching from parties, they are forming new groups that ease
their cognitive process of opinion formation and decision-making. We argue that Latin Ame-
rican politics is an ideal source for the study of such non-partisan associations. Whereas party
systems are generally weak (LUPU, 2016), the region has been divided by a series of cleavages
(ROSAS, 2010). Thus, we choose a topic that goes beyond the traditional State vs. market

in creating a safety net for the poorer strata of the population. Furthermore, we analyze this
issue from an understudied perspective: legislators’ self-declared attitudes in cross-national
comparative perspective.
We test two alternative hypotheses: descriptive representation and association to busi-
-
ring socio-demographic characteristics of traditionally dominant groups in the region’s socie-
ties – namely, the older, wealthy, well-educated, and male. We propose that belonging to such
identities reduces the perception that safety nets for the poorer strata are necessary. Second,
we test the relevance of self-association with business interests. The literature suggests a strong
-
cing (e.g., SANTOS et al., 2017). In this sense, legislators closer to these actors should focus more
on the defense of alternative issues, such as private property, low taxation, and protectionism.
These two tests allow verifying the assertiveness of each hypothesis, further than comparing the
weight of each of them in the formation of attitudes in national parliaments.


    

quo components. Furthermore, descriptive representation plays a stronger role in the beha-
vior of Latin American representatives than association with business interests. In brief, the
            

This paper is divided as follows. After this introduction, we present a brief discussion
of the literature that shows the decay of partisanship and the emergence of descriptive re-

Then, we present our research design, framing our hypotheses, detailing our data source, and
-
levant implications of this work and suggested future steps in the study of legislative behavior.
STATUS QUO CHARACTERISTICS OR PRIVATE SECTOR AFFINITY? EXPLAINING
PREFERENCES FOR SAFETY NETS IN LATIN AMERICAN LEGISLATURES
18
2. THEORY
Partisanship has long been portrayed in the literature as an important cue for legis-
lators’ behavior. However, parties have become increasingly more unresponsive around the
world. It led to lower loyalism (HOOGHE and STOLLE, 2005) even in countries with incre-
asing engagement in non-partisan organizations (KOVACHEVA, 2005; DALTON, 2008).
Indeed, it is possible that legislators and party elites purposefully engage in the weakening
of their parties’ membership basis. It is explained by the formation of alienated party oligar-
chies (MICHELS, 1915; COUTO, 2012; PANEBIANCO, 1994; MORENO and ALMEIDA, 2017;
GUEDES-NETO, 2018a) and the high costs of sustaining leading roles in parties with strong
democratic structures (SCARROW, 1994).
Nevertheless, this argument does not lead uniquely to the proposition that parties do
not act as cohesive institutions anymore. Even though many of them may have become un-
responsive to their loyalists, several of these organizations have realigned themselves to reach
 
NETO, 2015), or the median voter (DOWNS, 1957; EZROW et al., 2010; WARD et al., 2011).
In this sense, aggregate models of partisan behavior in legislatures still achieve relevant sig-

On the other side, an emerging literature has been evidencing the role of legislators as
individual actors with their own interests and networks. The recent growth of populist leaders
as Donald Trump and Geert Wilders functions as relevant examples (INGLEHART and NOR-
RIS, 2017; GUEDES-NETO, 2018b). Trump became president of the United States running as
an outsider in the traditional Republican Party (MACWILLIAMS, 2016). Wilders was a spee-
ch-writer of a well-established Dutch party, the Volkspartij voor Vrijheid en Democratie, but
gained prominence after founding his own far-right party (Partij voor de Vrijheid) (GUEDES-

and use them to gain representative power in their countries2.
In Latin America, the study of individual legislators’ behavior is even more relevant.
Party systems in the region are generally weak, especially in countries with low economic
development (LUNA and ZECHMEISTER, 2005). From the mid-1990s to the mid-2010s, a
quarter of the established parties in Latin America ceased to exist or became irrelevant (LUPU,
2016). Several others were created. Most of the remaining parties are not internally cohesive.
They tend to self-place themselves similarly in the left-right spectrum but highly disagree

2010). Deegan-Krause (2007) argues that it is due to the institutional setting, as voters choose
their representatives through open list systems. Thus, the candidate’s individual image and
linkage to his or her constituency become stronger cues than partisanship. In this scenario,
legislators may prefer to answer directly to their support bases rather than to their parties
2 Here, we do not argue that Trump’s constituency was not the same that has traditionally voted for the Re-
publican Party. However, as MacWilliams (2016) proposes, Trump has managed to activate a portion of the
electorate that was not attracted by insiders of the party.
19
TEORIAE SOCIEDADE Nº 26.2 - JULHO - DEZEMBRO DE 2018
(KEEFER and KHEMANI, 2009)3.
If some legislators do not necessarily respond to their own parties, which interests do
 -
siness interests and descriptive representation. Furthermore, we pick policies related to the
economy as our benchmark. We do so because of the relevance of economic voting in Latin
America (WEYLAND, 2003), as well as due to the stark income inequality in the region (GAS-
PARINI et al., 2011), and the recent pink tide, which has primed social protection policies
especially during the 2000s (CARNES and MARES, 2015). Furthermore, as our descriptive
-
ting jobs, and the government control over pensions are the ones that produce the strongest
cleavages in Latin American legislatures.
Dahm et al. (2014) argue that legislators can favor business preferences even when
dealing with a policy that may harm their own interests. This behavior is common worldwide
and frequently investigated by the students of lobbying and business interests. For instance,

        
majorly related to representing the interests of local industries. In the Brazilian case, Santos

aligned with the political agenda of the Confederação Nacional da Indústria, the major asso-
ciation of industrial interests in the country. Indeed, Brazilian lobbyists see the legislative as
their main target and perceive this strategy to be productive in 98.2% of the cases (SANTOS
et al., 2017).
Alternatively, many students of the legislative have aimed at varieties of descriptive re-
presentation. It involves many aspects. First, the literature has demonstrated that the average
age of legislators is generally high and, in many parliaments, has been increasing (STOCKE-
MER and SUNDSTRÖM, 2018). Whereas in some cases the lack of a generational turnover
favors the status quo, i.e. the ruling parties and elites (GOLOSOF, 2014), there is also evidence
that older and younger legislators hold similar policy preferences (KISSAU et al., 2012). Simi-
 
Atkeson and Carrillo (2007) demonstrated that the increase in the number of congresswomen
 
Franceschet and Piscopo (2008) argue that, at least in the Argentine case, descriptive and
substantive representation are not the same. They propose that the quota laws may have rein-

Most of these studies focus on issue ownership. They generally regard the representa-
tion exercised in areas that are directly related to the descriptive attributes of the legislator.
Indeed, certain groups tend to be focused on their policy domains also because they are margi-
3 The reader should have in mind that whereas this generalization is adequate to comparative regional studies,
       
works.
STATUS QUO CHARACTERISTICS OR PRIVATE SECTOR AFFINITY? EXPLAINING
PREFERENCES FOR SAFETY NETS IN LATIN AMERICAN LEGISLATURES
20
nalized by the status quo, which leads them to work only on the areas stereotypically linked to
their identities (SCHWINDT-BAYER, 2006). However, it does not mean that they do not hold
their own policy preferences in other domains. Indeed, such marginalization may lead them
to polarize against the status quo in areas other than the ones they are typically involved with.
3. RESEARCH DESIGN
-
rican Elites Database from the University of Salamanca (PELA-USAL). This database4 gathers
the results of close-ended interviews carried out by members of the research team with Latin
American legislators (ALCÁNTARA, 2012; GARCÍA et al., 2013). The interviews are conduc-
   
chosen from political parties with legislative representation. The sample is proportional to
political parties’ sizes. In the questionnaires, a wide range of issues is included, such as demo-
cracy, foreign policy, political parties, ideology, political career, social issues, and Statism and
public policies. This has led to works about a considerable variety of issues, from polarization
(RIVAS, 2008) to political careers (MARTÍNEZ ROSÓN, 2009) or support for democracy
(BOHIGUES, 2018).
In this paper, we analyze legislators’ attitudes towards public policies. We selected the
most recent studies for 17 countries, which are those where the relevant questions are availa-
ble: Argentina (09-13; 11-15), Peru (11-16), Paraguay (13-18), Honduras, (14-18), Costa Rica
(14-18), Colombia (14-18), Chile (14-18), Panama (14-19), Mexico (15-18), El Salvador (15-18),
Uruguay (15-20), Bolivia (15-20), Guatemala (16-20), Dominican Republic (16-20), Venezuela
(16-21), Ecuador (17-21), and Nicaragua (17-22).
Our questions of interest to measure and analyze legislators’ attitudes towards public
policies5 are the following: (1) The State, instead of the private sector, should be the owner
of the most important companies and industries in the country; (2) The State, rather than
individuals, should be the main responsible for ensuring the welfare of the people; (3) The
State, rather than the private company, should be the main responsible for creating jobs; (4)
;
(5) The State, rather than the private sector, should be the main responsible for providing
retirement pensions; (6) The State, rather than the private sector, should be the main res-
ponsible for providing health services; (7) The State, rather than the private sector, should be
the main responsible for providing university education; (8) The State must implement pu-
blic policies to reduce inequalities between men and women; (9) The State must implement
. For each
4 This free public-access database belongs to the University of Salamanca. Find more information at http://
americo.usal.es/oir/elites/eliteca.htm or by contacting the authors.
5 Here, we adopt a broad concept of public policy, that is, the role of the State in intervening in certain policy

in the sample.
21
TEORIAE SOCIEDADE Nº 26.2 - JULHO - DEZEMBRO DE 2018
of the nine items, legislators are given a scale that ranges from 1 to 7, where 1 means strongly
disagrees and 7 strongly agrees.
-
blic policies have among legislators in Latin America. To do so, we begin with the HJ-Biplot
technique. It assesses the levels of dis/agreement with the aforementioned items, which Par-
liaments are comparatively more receptive to a more active role of the State in the economy
and society, and identify regional clusters. Second, we identify which policies are the most
controversial, that is, that do not generate consensus among the political elites. As we will see
later, some policies are, at least rhetorically, accepted, with high levels of agreement and low
levels of disparity in the answers – as measured by the standard deviation. Third, we analyze
what drives support to these far-from-consensus items, building on the theoretical framework
presented in the previous section. Namely, we propose models that have as their independent
variables the socio-demographic characteristics (i.e., dummy variables for status-quo iden-

dependent variables.
1, is that non-status quo legislators are more pro-
ne to support an active role of the State in the creation of safety nets for the poorer strata of
the population. We consider two assumptions to propose this hypothesis. First, our expecta-
tion derives from the Rawlsian notion of the veil of ignorance. Once non-status quo legislators

policies that will assure minimum life standards to everyone. Second, we assume that status-
quo politicians are male, older, well-educated, and wealthier thus reproducing the typical
stereotype of the Latin American economic elite.
As we also suggest that business interests shape legislators’ policy preferences, our
second hypothesis, H2, regards that representatives who either hold high-level jobs in the

business associations are more prone to go against the economic policies that increase the
size of the State or favor the working class. It should be noticed that it is not the same as to
argue that these legislators are classical liberal or, in other words, favorable toward a mini-
mum State. It simply means that they do not want tax money and State capacity to be directed

although it is not our objective to test, favor protectionist policies to secure local markets and
facilitate exports. In our case, we consider the before-mentioned policies related to the public

It is noteworthy mentioning that, in these analyses, we include three factors undou-
btedly important in Latin American politics, but have a secondary role in this paper: coun-
      

we mainly focus on sociodemographic characteristics of the legislators, we have decided to
include these other two variables as controls and as proxy mechanisms for partisanship: the
STATUS QUO CHARACTERISTICS OR PRIVATE SECTOR AFFINITY? EXPLAINING
PREFERENCES FOR SAFETY NETS IN LATIN AMERICAN LEGISLATURES
22

to belong to the opposition or to the government or presidential coalition. We expect that
left-leaning respondents are more willing to support a more active State and that legislators
of the opposition are not, given that, when in opposition, legislators should prefer a rather
passive role of the State.
We highlight that the latter two variables are present simply to control for possible

government-opposition self-placement in day-to-day politics. However, these general proxies
shall reduce any noise remaining from party bonds in the selected parliaments. As we have
argued before, we aim to analyze the impact of descriptive representation, namely sociodemo-
graphic factors, and business coalitions.
4. REGIONAL CLUSTERS AND (NON) CONTROVERSIAL POLICIES
The descriptive analysis will make use of the factor statistical method HJ-Biplot, which
is a graphic representation of multivariate data (ALCÁNTARA and RIVAS, 2007). If a scatter
plot shows the distribution of two variables, a Biplot allows the representation of cases and
three or more variables simultaneously. In the HJ-Biplot, variables are represented by vectors
and cases by points (countries in the case of our work). For the representation of these cases
and variables, the HJ-Biplot runs a factor analysis, which serves to frame this variables-cases
graphical representation.
In the resulting graph, we can see the variability of each variable by observing the len-
gth of the vectors and the angle they form with the factorial axes. Thus, the longer the vector
(variable) and the smaller the angle it forms with the factor axis that represents it (axis/factor
1 or axis/factor 2), the higher the variance explained by this variable. It is also possible to
know the correlation between variables by observing the angle they form: if the angle is less
than 90º, there is a positive correlation; if it is greater than 90º, there is a negative correlation;
and if it approaches 90º, there is no correlation. The relationship between cases and variables
(vectors) is obtained by the perpendicular projection of the cases on the vectors. Proximity
between cases is interpreted as a similarity between the cases and eventually leads to the cre-
ation of clusters. In short, HJ-Biplot allows simultaneous visualization, in a single graph, both
variables (nine public policies) and the relationship (correlations) between themselves, and
the position of each case (17 countries6) with respect to those variables.
The HJ-Biplot has been used in recent works, such as Alcántara and Rivas (2007),
Rivas et al. (2010), Alcántara (2013), and Bohigues (2018). It is due to these advantages and
simplicity that we decided to opt for this technique. In essence, it synthesizes in a single graph
the information coming from our nine questions and 17 countries of interest (Figure 1).

6 In this part of the paper we use countries means, since here the case are countries and not individual legis-
lators.
23
TEORIAE SOCIEDADE Nº 26.2 - JULHO - DEZEMBRO DE 2018
than 90º. For example, the relationship between health services and cultural and ethnic
groups inequality is the lowest, since the angle they form is nearer to 90º, but there is still a


parliaments most reluctant to an active role of the State; on the contrary, Bolivia and Uruguay
are the most willing to support active public policies in areas such as inequality, health, and
creation of jobs. The rest of the countries fall between these two apparent groups.
Figure 1. Graphic representation of the countries and
support for public policies of Latin American legislator
based on HJ-Biplot analysis
Source: own elaboration based on PELA-USAL and HJ-Biplot.
As it has been said, another advantage of HJ-Biplot is the possibility of creating clus-
ters form their biplot coordinates (Figure 2). The advantage of cluster analysis is that it allows
homogeneous groups to be established in such a way that individuals belonging to the same
group are similar, that is, their values in each of the variables used in the analysis are similar.
STATUS QUO CHARACTERISTICS OR PRIVATE SECTOR AFFINITY? EXPLAINING
PREFERENCES FOR SAFETY NETS IN LATIN AMERICAN LEGISLATURES
24
Figure 2. Dendrogram of clusters of countries according to
public policies support

Source: own elaboration based on PELA-USAL and
HJ-Biplot coordinates (Linkage: Ward).
Bolivia and Uruguay form, indeed, a cluster – one that is rather favorable toward safety
nets. Additionally, Colombia, Guatemala, Paraguay, Chile, and Peru compose another cluster,
-
tremes: Nicaragua, Mexico, Costa Rica and Argentina, with medium-low values for gender,
socioeconomic and indigenous inequalities, and rather low for the rest; and El Salvador, Ecua-
dor, Venezuela, Honduras, Dominican Republic and Panama, with medium-high values for all
nine policies. It is worth bearing in mind that Bolivia and Uruguay have left-leaning legislative
majorities in the legislatures included in this paper: the governing Movement towards Socia-
lism (Movimiento al Socialismo) and Broad Front (Frente Amplio)7, respectively. Overall,

-
cies according to their degree of agreement, that is, their degree of consensus. Figure 3 gathers
the aggregate means (in the whole database with the PELA-USAL studies here included) of
each item and their respective standard deviations (SD).
 Partido Indepen-
diente) and Popular Unity (Unidad Popular).
25
TEORIAE SOCIEDADE Nº 26.2 - JULHO - DEZEMBRO DE 2018
Figure 3. Support for public policies in Latin American
Parliaments (merged database)




 





- Ordered from higher to lower mean and SD.
Source: own elaboration based on PELA-USAL.
Undoubtedly, the public ownership of the most important industries and companies,
the creation of jobs and the management of retirement pensions are not only the most con-
troversial (higher SDs), they also have the lowest levels of support. Additionally, and maybe
not by chance, these three policies are directly related with the guarantee of safety nets for

relevant cleavage in Latin America.
5. RESULTS

run three models for each dependent variable. Model 1 tests status-quo variables: age, male (di-
chotomous, 1 for male), salary, and education, plus an interaction of these four variables, which


 
and perceived trustworthiness of employers’ organizations, plus an interaction of these two, which
-

-

tests all these variables, plus our two control variables, ideology (1-10 scale, being 1 left) and oppo-
sition (dichotomous, 1 for opposition). Table 1 presents the three models for each item.
STATUS QUO CHARACTERISTICS OR PRIVATE SECTOR AFFINITY? EXPLAINING
PREFERENCES FOR SAFETY NETS IN LATIN AMERICAN LEGISLATURES
26
Table 1. Multivariate linear OLS regressions explaining
legislators’ preferences toward public ownership of firms,
the role of the State in the creation of jobs and assurance
of retirement benefits
Public ownership Creation of Jobs Retirement
(1) (2) (3) (1) (2) (3) (1) (2) (3)
Age .009 .01+ 006 .005 .01** .01***
Male -.22 -.01 -.16 -.1 .06 .14
Salary -.44*** -.16 -.29*** -.24** -.20** -.14
Education -.26*** .19*** -.32*** -.30*** -.09+ -.06
Status quo 0-.001** 0-.30*** 0-.001+
Businessman -1.03+ -.74 .524 .58 -.465 -.45
Employers’ Org -.55*** -.34*** -.004 .03 -.079 -.05
 .21 .18 -.177 -.20 .171 .19
Ideology -.27*** -.04 -.05**
Opposition -.67*** -.57*** -.27**
(Constant) 4.91*** 4.23*** 6.53*** 6.62*** 4.24*** 6.85*** 5.71*** 5.25*** 5.73***
Adjusted R2.209 .177 .341 .208 .138 .231 .158 .133 .167


Source: own elaboration based on PELA-USAL.
-
sions. Older legislators are favorable toward the State’s role in assuring this policy. Similarly,


-

support for these policies.
The other two variables that compose the status quo category do have stronger and
              
          
(public ownership and retirement) or become slightly lower (creation of jobs). It suggests


want the State to be active in the creation of jobs but hold dispersed opinions when it comes

Second, education is a parallel predictor, since it works the same way: the higher the
level of studies, the lower the support for the safety net. This is in sharp contrast with evidence
from Latin American citizens themselves (DURAKIEWICZ, 2018). Although at the mass level
27
TEORIAE SOCIEDADE Nº 26.2 - JULHO - DEZEMBRO DE 2018
education means more acceptance of redistributive policies, the opposite becomes true when
the preferences of legislative elites are assessed. However, education, just as salary, has no

case, again, that highly educated legislators have closer ties to the business sector.
-
i=0 is rejected with the generous 0.1 level, while
           
quo matter for the rejection of social safety nets. However, as earlier said, when considered
separately, not all status-quo variables perform in the same way: education and salary have

predictor.
-

-
sions, contrary to what the literature has usually stated. As Table 1 puts it, except for public
ownership’s model 2 , businessmen are neither more nor less supportive of public policies,

On the contrary, trusting employer’s organizations serves as a strong predictor for the

legislator with employers’ organizations, the less she or he favors the State’s control over com-


of the agenda of business interests8.
Regarding our two control variables, results are as expected. Left-leaning positions are


State’s role in the creation of jobs: both leftists and rightist support/reject in similar levels of
dispersion this public policy, meaning it is beyond the left-right divide.
When it comes to the government vs.
-

they prefer a rather passive State in the economy due to this possible instrumental motivation.
It is important to bear in mind that the category Opposition includes left, center and right leg-


each group of variables by themselves. In all the cases, variables related to the status quo had a

8 -

STATUS QUO CHARACTERISTICS OR PRIVATE SECTOR AFFINITY? EXPLAINING
PREFERENCES FOR SAFETY NETS IN LATIN AMERICAN LEGISLATURES
28
R2
descriptive representation is more relevant than the relationship of legislators with business

our H12

that H1, the status quo descriptive representation hypothesis, is a better explanation for the
rejection of safety nets to the poorer strata of the population.

-


one in which salary remains a strong predictor in the full model (3). Retirement pensions are,


that can explain levels of support.
6. DISCUSSION AND CONCLUSION
This paper reviewed the literature of partisanship to suggest, in agreement with pre-
vious works, that other elements may function as better predictors of legislators’ identity
when it comes to the cognitive process of preferences’ formation. Following this assumption,
-
ters. Further than considering gender and age as commonly seen in the literature, we added
sociodemographic variables for the level of education aand salary. We suggested that these
  
-

shapes policy preferences in the legislative. These two hypotheses were tested based on one
of the strongest ideological cleavages in Latin America: the role of the State in guaranteeing a
safety net to the poorer strata of the population – an argument derived from the Rawlsian view
of social security.
Before running the relevant regressions, we used the data to make a general assess-
ment of the distribution of these policy preferences in Latin America. We showed that there

more favorable toward a safety net. The opposite is true for Colombia, Guatemala, Paraguay,
Chile, and Peru. Indeed, it evidences something already expected once Bolivia and Uruguay

-


regional cleavage and reassuring their relevance to our study.
29
TEORIAE SOCIEDADE Nº 26.2 - JULHO - DEZEMBRO DE 2018
When it comes to the regressions of interest to our test of hypotheses, the proposition
     




to their proximity with the age of retirement, thus corroborating to an adjusted descriptive re-
presentation hypothesis. However, when it comes to the interactive term, our results support
the hypothesis that older, male, wealthy, and highly educated legislators tend to hold policy
preferences against the provision of safety nets to the poorer strata of the population – even

We also highlight that higher levels of education are usually portrayed in the litera-
ture as generating higher favorability toward redistributive policies at least when it comes to
-
ring our results to the existing literature suggests that highly educated voters may be better
represented by lowly educated legislators (and vice versa) at least when it comes to policy
preferences toward safety nets. This topic should be further studied in the future.

the private sector lead to the rejection of the State’s role in assuring social security mechanis-
 
    
 
of business interests in Latin America at least in regard to the formation of cognitive cues for
policy preferences. We highlight that these results do not provide an argument against the role

preferences and the behavior of legislators in the policy-making process. In this sense, we su-
ggest further studies that investigate when the ideology of legislative elites matters.
Finally, our control variables evidenced two issues that could be addressed by other
students of policy preferences in the future. First, left- and right-leaning legislators converge
in their preferences toward the State’s role in the creation of jobs. We assume that it results
-
le some legislators may be thinking about lower taxes, others may have considered protectio-
nist policies. Second, opposition representatives tend to go against the provision of safety nets
even when ideology is controlled. As we suggested, it may be explained by their motivation to
see a passive government, as to allow future electoral gains. However, we prefer not to theo-
rize on this issue once it escapes the scope of this paper. It would be interesting to see studies
addressing this issue.
STATUS QUO CHARACTERISTICS OR PRIVATE SECTOR AFFINITY? EXPLAINING
PREFERENCES FOR SAFETY NETS IN LATIN AMERICAN LEGISLATURES
30
7. BIBLIOGRAPHY
ALCÁNTARA, M. 2012. . Madrid: Tecnos.
ALCÁNTARA, M. (2013). “Los retos actuales de la política en América Latina”. Revista Mexicana
De Análisis Político y Administración Pública, v. 2 n. 2, p. 9-30. ALCÁNTARA, M. and RI-
VAS, C. 2007. “Las dimensiones de la polarización partidista en América Latina”. Política
y Gobierno, v. 14, n. 2, p. 349-390.
-
Politics and Gender, v. 3, n. 1, p. 79-101.
BOHIGUES, A. “El apoyo a la democracia en los parlamentos latinoamericanos. Análisis del de-
mócrata ambivalente”. Revista de Ciencia Política, v. 38, n. 1, p. 51-82.

employment vulnerability, income, and preferences for social protection in Latin Ameri-
ca”. Politics and Society, v. 43, n. 4, p. 525-550.
COUTO, C. 2012. “Oligarquia e Processos de Oligarquização: o Aporte de Michels à Análise Política
Contemporânea”. Revista de Sociologia e Política, v. 20, n. 44, p. 47-62.
DALTON, R., 2008. “Citizenship norms and the expansion of political participation”. Political Stu-
dies, vol. 56, p. 76-98.

Public Choice, v. 159, p. 63-82.
DEEGAN-KRAUSE, K. 2007. “New dimensions of political cleavage”. In: DALTON, R., KLINGE-
MANN, H-D. (Org.), Oxford Handbook of Political Behavior. Oxford: Oxford University
Press, p. 538-556.
DOWNS, A. 1957. “An Economic Theory of Political Action in a Democracy”. Journal of Political
Economy, v. 65, p. 135-150.
DURAKIEWICZ, P. 2018. “Reducing Inequality in the Americas: What Factors Predict Public
Support for Redistribution?”. AmericasBarometer Insights Series, 132.
EZROW, L., DE VRIES, C., STEENBERGEN, M., EDWARDS, E. 2010. “Mean voter representa-
tion and partisan constituency representation: Do parties respond to the mean voter posi-
tion or to their supporters?”. Party Politics, v. 17, n. 3, p. 275-301.
FRANCESCHET, S., PISCOPO, J. 2008. “Gender quotas and women’s substantive representa-
tion: Lessons from Argentina”. Politics and Gender, v. 4, n. 3, p. 393-425.
GARCÍA, F.; MATEOS, A. and RIVAS, C. 2013. “Veinte años de élites parlamentarias en américa
latina (1994-2014)”. Revista De Las Cortes Generales, v. 89, p. 135-174.
GASPARINI, L., CRUCES, G., TORNAROLLI, L. 2011. “Recent trends in income inequality in La-
tin America”. Economia, v. 11, n. 2, p. 147-190.
GOLOSOV, G. 2014. “Growing Old without Grace: Electoral Authoritarianism and the Age Com-
position of Russia’s Regional Legislative Assemblies”. Representation, v. 50, n. 4, p. 509-
31
TEORIAE SOCIEDADE Nº 26.2 - JULHO - DEZEMBRO DE 2018
526.
GUEDES-NETO, J. 2015. “A Reversed Double Movement in Brazil: the (Controversial) Evolution
of the Partido dos Trabalhadores’ Education Policy since 1980”. Journal of International
Studies, v. 8, n. 2, p. 139-154.
GUEDES-NETO, J. 2018a. “Transformação Intrapartidária? Um estudo sobre as percepções de
Revista de Sociologia e Política, v. 26, n. 68.
GUEDES-NETO, J. 2018b. “On Social Capital, Creative Class, Emancipative Values, and the Ri-
sing Far-Right Populism”. Conjuntura Internacional, v. 15, n. 1, p. 1-12.
HAWKINS, K., MORGENSTERN, S. 2010. “Ideological cohesion of political parties in Latin
America”. In: KITSCHELT, H., HAWKINS, K., LUNA, J., ROSAS, G., ZECHMEISTER, E.
(Org.), Latin American Party Systems. Cambridge: Cambridge University Press, p. 145-
168.
HOOGHE, M; STOLLE, D. 2005. “Youth organisations within political parties: political recruit-
ment and the transformation of party systems”. In: FORBRIG, J. (Org.) Revisiting youth
political participation: challenges for research and democratic practice in Europe. Es-
trasburgo: Council of Europe.
INGLEHART, R., NORRIS, P. 2017. “Trump and the populist authoritarian parties: the silent re-
volution in reverse”. Perspectives on Politics, v. 15, n. 2, p. 443-454.
KEEFER, P., KHEMANI, S. 2009. “When do legislators pass on pork? The role of political parties
American Political Science Review, v. 103, n. 1, 2009, p.
99-112.
KISSAU, K., LUTZ, G., ROSSET, J. 2012. “Unequal representation of age groups in Switzerland”.
Representation, v. 48, n. 1, p. 63-81.
KOVACHEVA, S. 2005. “Will youth rejuvenate the patterns of political participation?”. In: FOR-
BRIG, J. (Org.) Revisiting youth political participation: challenges for research and de-
mocratic practice in Europe. Estrasburgo: Council of Europe.
LUNA, J., ZECHMEISTER, E. 2005. “Political representation in Latin America: a study of elite-
mass congruence in nine countries”. Comparative Political Studies, v. 38, n. 4, p. 388-416.
LUPU, N., 2016. Party brands in crisis: partisanship, brand dilution, and the breakdown of po-
litical parties in Latin America. Cambridge University Press.
MACWILLIAMS, M. 2016. “Who Decides When The Party Doesn’t? Authoritarian Voters and the
Rise of Donald Trump”. PS: Political Science & Politics, v. 49, n. 4, p. 716-721.
MARTÍNEZ ROSÓN, M. 2009. “La ambición política en situaciones adversas: contextos institu-
cionales y personales”. Revista de Estudios Políticos, v. 146, p. 113-148.
MICHELS, R., 1915. Political Parties: a Sociological Study of the Oligarchical Tendencies of Mo-
dern Democracy. Traduzido por Eden & Cedar Paul. Nova Iorque: Hearst’s International
Library.
MORENO, R.; ALMEIDA, A. 2017. “Quando jovens ativistas do hip hop encontram a política par-
STATUS QUO CHARACTERISTICS OR PRIVATE SECTOR AFFINITY? EXPLAINING
PREFERENCES FOR SAFETY NETS IN LATIN AMERICAN LEGISLATURES
32
tidária”. Revista de Sociologia e Política, vol. 25, n. 61, p. 5-29
OLSON, M., 2002. The logic of Collective Action: Public Goods and the Theory of Groups. Cam-
bridge: Harvard University Press.
PANEBIANCO, A. 1994. “Poder y Organización en los Partidos Políticos I”. Revista Mexicana de
Ciencias Políticas y Sociales, v. 39, n. 156, p. 13-29.
RIVAS, C. 2008. “The dimensions of Polarization in Parliaments”. In: Manuel Alcántara ez (ed.).
Politicians and Politics in Latin America. Boulder: Lynne Rienner Publishers, p. 139-160.
RIVAS, C.; VICENTE, P. and SÁNCHEZ, F. 2010. ”La educación como elemento de calidad de los
políticos latinoamericanos“. Política y Gobierno, v. 17, n. 2, p. 279-319.
ROSAS, G., 2010. “Issues, ideologies, and partisan divides: imprints of programmatic structure on
Latin American Legislatures”. In: KITSCHELT, H., HAWKINS, K., LUNA, J., ROSAS, G.,
ZECHMEISTER, E. (Org.), Latin American Party Systems. Cambridge: Cambridge Uni-
versity Press, p. 70-95.
SAMUELS, D.; SHUGART, M. 2010. Presidents, Parties, and Prime Ministers: How the Separa-
. Cambridge: Cambridge Univer-
sity Press.
SANTOS, M., SILVA, M., FIGUEIREDO FILHO, D., ROCHA, E. 2015. “Financiamento de cam-
panha e apoio parlamentar à Agenda Legislativa da Indústria na Câmera dos Deputados”.
Opinião Pública, v. 21, n. 1, 2015, p. 33-59.
-
Textos para Discussão (IPEA), n. 2,334, p. 1-66.

Memberships”. European Journal of Political Research, v. 25, p. 41-60.
SCHWINDT-BAYER, L. 2006. “Still Supermadres? Gender and the Policy Priorities of Latin Ame-
rican Legislators”. American Journal of Political Science, v. 50, n. 3, p. 570-585.
STOCKEMER, D., SUNDSTRÖM, A, 2018. “Age representation in parliaments: Can institutions
pave the way for the young?” European Political Science Review, v. 10, n. 3, p. 467-490.
WARD, H., EZROW, L., DORUSSEN, H. 2011. “Globalization, party positions, and the median
voter”. World Politics, v. 63, n. 3, p. 509-547.
WEYLAND, K. 2003. “Economic voting reconsidered: Crisis and charisma in the election of Hugo
Chávez”. Comparative Political Studies, v. 36, n. 7, p. 822-848.
WOOD, D. 1987. “The Conservative Member of Parliament as Lobbyist for Constituency Econo-
mic Interests”. Political Studies, v. 35, p. 393-409.
... All models are nested within legislatures (country-year) and clustered for standard errors at the same level. The full results are presented in Appendix 1 Table 3, model 10. liberalisation (Guedes-Neto & Bohigues, 2018). In most Latin American countries, the average civil service legislator leans left. ...
Article
Full-text available
The failure of political parties to adequately voice the preferences of their constituents leads voters to look for alternative heuristics when selecting their favourite candidates. Descriptive representation is one of them. Yet, most scholars focus only on the social characteristics of representatives. Is it also possible to talk about a workplace descriptive representation? In this paper, I rely on a longitudinal survey conducted with 2,527 legislators from 18 Latin American countries to assess whether the preferences of those representatives with a background in the public service differ from those held by other congressmen. The results show that former public employees adequately represent their professional group only when they self-place in the left or highly identify with the civil service. Yet, when they do not, their preference for the maximisation of the state's budget and power is even lower than that of rightist legislators who have never worked for government.
Article
Full-text available
What explains the attitudes of Latin American political elites towards same-sex marriage, abortion, and drug legalization? In recent years, several countries have engaged in political disputes centred on these three issues. Scholarship has explored the topic from different perspectives, ranging from theories of democracy to voters’ political behaviour. Nonetheless, the perspective of political elites is still understudied. We use data from elite surveys in sixteen Latin American countries to address this question. First, we find significant differences between and within countries and political parties, meaning these issues are far from consensual. Second, we analyse the effects of ideology, religion, and socio-demographic characteristics. We confirm that ideology and religiosity are key predictors, but a deeper analysis shows that whereas those on the right and religious are strongly op-posed to these issues, those on the left and less religious are not as strongly in favour.
Article
Full-text available
What explains the attitudes of Latin American political elites towards same-sex marriage, abortion, and drug legalization? In recent years, several countries have engaged in political disputes centred on these three issues. Scholarship has explored the topic from different perspectives, ranging from theories of democracy to voters’ political behaviour. None-theless, the perspective of political elites is still understudied. We use data from elite surveys in sixteen Latin American countries to address this question. First, we find significant dif-ferences between and within countries and political parties, meaning these issues are far from consensual. Second, we analyse the effects of ideology, religion, and socio-demographic characteristics. We confirm that ideology and religiosity are key predictors, but a deeper analysis shows that whereas those on the right and religious are strongly op-posed to these issues, those on the left and less religious are not as strongly in favour.
Article
Full-text available
SUMARIO: I.- Antecedentes y estado de la cuestión. II.- Objetivos y diseño metodológico. III.- Decisiones metodológicas a lo largo de la investigación. A) Decisiones metodológicas derivadas del cuestionario. B) Decisiones metodológicas derivadas de la unidad de análisis. IV.- Contribución académica. V.- Bibliografía utilizada y especializada.
Article
Full-text available
Putnam (1995), Florida (2002) and Welzel (2013) have at least one common argument: social capital, in the way we knew it, is dying. According to new theories, it is being replaced by a different mode of interaction, marked by weak-ties relationships and benign individualism. If that is true, why is far-right populism becoming so strong in Western democracies? This paper does not aim at creating a new theory or at presenting a final answer for this question but rather at pointing out at why Richard Florida’s and Christian Welzel’s theories did not anticipate the rise of a 21st-century mode of populism. As a core concluding element, I highlight the unforeseen triggering role played by increasing emancipative values in the mobilization of the now far-right electorate
Article
Full-text available
Este artículo tiene como objetivo analizar el perfil de la élite legislativa latinoamericana que no apoya incondicionalmente la democracia: los demócratas ambivalentes. Aunque diversos estudios han entrado a analizar el papel de la élite en los procesos democráticos, no se conoce cómo son los demócratas ambivalentes de este selecto grupo político. Así, este análisis, sin dejar de lado el contexto ni las diferencias entre países, permitirá conocer los valores y actitudes característicos de los diputados que, en determinadas circunstancias, apoyarían gobiernos autoritarios. Los resultados muestran que, al contrario de lo establecido por la literatura, no son marcadamente radicales, desconfiados ni intolerantes.
Article
Full-text available
O estudo analisa dados de uma pesquisa do tipo survey, realizada com representantes de entidades públicas ou privadas cadastrados na Câmara dos Deputados. As questões dizem respeito a três dimensões do lobbying: a recente profissionalização da atividade, estratégias e repertórios próprios desses grupos e, por fim, as percepções dos respondentes sobre sua influência no processo decisório. Os resultados apontam para uma tendência de profissionalização da atividade no Brasil; para um repertório de ações bastante amplo, mas centrado na troca e na disseminação de informações como principal ativo dos grupos de interesses; e para uma visão otimista a respeito da sua capacidade de influência parlamentar.
Article
Full-text available
Growing up taking survival for granted makes people more open to new ideas and more tolerant of outgroups. Insecurity has the opposite effect, stimulating an Authoritarian Reflex in which people close ranks behind strong leaders, with strong in-group solidarity, rejection of outsiders, and rigid conformity to group norms. The 35 years of exceptional security experienced by developed democracies after WWII brought pervasive cultural changes, including the rise of Green parties and the spread of democracy. During the past 35 years, economic growth continued, but virtually all of the gains went to those at the top; the less-educated experienced declining existential security, fueling support for Populist Authoritarian phenomena such as Brexit, France’s National Front and Trump’s takeover of the Republican party. This raises two questions: (1) “What motivates people to support Populist Authoritarian movements?” And (2) “Why is the populist authoritarian vote so much higher now than it was several decades ago in high-income countries?” The two questions have different answers. Support for populist authoritarian parties is motivated by a backlash against cultural change. From the start, younger Postmaterialist birth cohorts supported environmentalist parties, while older, less secure cohorts supported authoritarian xenophobic parties, in an enduring intergenerational value clash. But for the past three decades, strong period effects have been working to increase support for xenophobic parties: economic gains have gone almost entirely to those at the top, while a large share of the population experienced declining real income and job security, along with a large influx of immigrants and refugees. Cultural backlash explains why given individuals support Populist Authoritarian movements. Declining existential security explains why support for these movements is greater now than it was thirty years ago.
Article
Se pretende mostrar la utilidad de los instrumentos del análisis organizativo (en la variante de la aproximación “conflictual”) para el estudio de la política, procediendo en tres tiempos: 1. La estadística de la organización: rol de los “fines” organizativos, naturaleza del poder, vínculo impuesto por la estructura organizativa, características de las relaciones entre organización y ambiente; 2. La dinámica organizativa: génesis de las organizaciones, de la institucionalización y del cambio organizativo; 3. Las organizaciones políticas propiamente dichas. En este trabajo se revisan algunos aspectos relacionados con el primer punto.
Book
Why have so many established political parties across Latin America collapsed in recent years? Party Brands in Crisis offers an explanation that highlights the effect of elite actions on voter behavior. During the 1980s and 1990s, political elites across the region implemented policies inconsistent with the traditional positions of their party, provoked internal party conflicts, and formed strange-bedfellow alliances with traditional rivals. These actions diluted party brands and eroded voter attachment. Without the assured support of a partisan base, parties became more susceptible to short-term retrospective voting, and voters without party attachments deserted incumbent parties when they performed poorly. Party Brands in Crisis offers the first general explanation of party breakdown in Latin America, reinforcing the interaction between elite behavior and mass attitudes.
Article
Middle-aged to senior men of the ethnic majority and higher income groups are generally overrepresented in parliaments. While research on group representation has examined issues of gender, economic standing, and, more recently, ethnicity, few studies examine age groups. We argue that the design of political institutions influences the share of young adults in parliaments across nations and hypothesize that the electoral system type, age candidacy requirements, and quotas influence the share of younger deputies in national parliaments. Analyzing an original data set with a global cross-national sample, we find that proportional representation and giving candidates the right to stand in elections as early as possible (i.e. at the age of 18) matter. In contrast, quota provisions for youths are currently too selectively applied to increase the percentage of young deputies in parliament.