Populism and policy change. The reform of the Ecuadorian 
mining sector under Rafael Correa between 2007 and 2010+
PAÚL CISNEROS*
Instituto de Altos Estudios Nacionales 
Centro de Gobierno y Administración Pública 
paul.cisneros@iaen.edu.ec y pcisneros@ucdavis.edu
https://doi.org/108800/rcpg.201802.004
Abstract
This article investigates the influence of populism in a process of policy change. The case of 
the mining law approved by the government of Rafael Correa between 2007 and 2010 is 
studied using the Advocacy Coalition Framework and discourse network analysis to study the 
structure of the narratives used by actors during the process of reform. The article uses an ori-
ginal dataset built from newspaper articles, which is analyzed using computational techniques. 
There are two main findings. First, the populist strategy that exploits emerging policy images to 
reduce institutional friction motivated reformist and incumbent coalitions to expand the con-
flict to promote their policy preferences. Second, the populist strategy deployed by president 
Correa reduced the degree of openness and diversity of the policy subsystem beyond what was 
expected in the policy process.
Keywords: social networks, policy change, populism, Advocacy Coalition Framework, mining 
law, Ecuador.
+ Recibido el 13 de marzo de 2019; aceptado el 20 de enero de 2020.
 I would like to thank Christopher Weible, Daniel Nohrstedt, Manuel Fischer and Mark Lubell, and 
two anonymous reviewers for comments on earlier versions of this paper. Naturally, I remain respon-
sible for any mistakes still present. Ecuador’s Secretary for Higher Education, Science Technology and 
Innovation (SENESCYT) financed this research through grant 27017.
* Doctor en Ciencias Sociales con especialidad en Estudios Políticos por la Facultad Latinoamericana 
de Ciencias Sociales FLACSO Sede Ecuador. Docente de políticas públicas del Centro de Gobierno y 
Administración Pública del Instituto de Altos Estudios Nacionales.
Revista de Ciencia Política y Gobierno, 5(9), 2018, 79-103
80 Revista de Ciencia Política y Gobierno, 5(10), 2018
Populismo y cambio de política. La reforma del sector minero ecuatoriano 
durante el gobierno de Rafael Correa entre 2007 y 2010 
Resumen
Este artículo analiza la influencia del populismo en el proceso de cambio de las políticas 
públicas. Para esto, utiliza el caso de la reforma a la ley minera realizada durante el gobierno de 
Rafael Correa entre el 2007 y el 2010. Se estudia la estructura del discurso que los actores de 
la política minera utilizan para incidir en el proceso de reforma a partir del Advocacy Coalition 
Framework y del análisis de redes discursivas. El artículo utiliza técnicas computacionales para 
analizar redes sociales. Los hallazgos principales del artículo son dos. En primer lugar, la estra-
tegia populista utilizada por el presidente Correa, que explora imágenes de política pública 
emergentes para reducir la fricción institucional, influyó en que tanto la coalición reformista 
como aquella defensora del status quo, expandieran el conflicto para posicionar sus preferen-
cias de política pública. Además, la estrategia populista de Correa redujo el nivel de apertura 
y diversidad del subsistema de política pública más allá de lo que se esperaba en el proceso de 
reforma de la política pública.
Palabras clave: redes sociales, cambio de política pública, populismo, Marco de Coaliciones de 
Causa, ley de minería, Ecuador.
 Paúl Cisneros / The reform of the Ecuadorian mining sector under Rafael Correa between 2007 and 2010 81
Introduction
During the past decade, several governments in Latin America reformed the 
legal framework of their mining sector to increase extractive rents during the 
super-cycle of commodities initiated in 2002 (Haarstad, 2012; Nem Singh 
& Bourgouin, 2013). In some cases, the initiatives of sectoral reform inter-
acted with system-wide processes of regime change that saw the re-emergence 
of populism at the center stage of politics (see Cisneros, 2011). In Ecuador, 
the regime headed by Rafael Correa (2007-2016) carried out a comprehensive 
reform of the mining sector in a local context characterized by an increase in 
social spending and the radicalization of demands for institutional innovations 
to deal with pressing social and environmental issues in the sector.
Current literature on recent mining policy reforms in Latin America 
is largely based on the idea that existing political institutions reinforce and 
expand extractive sectors’ activities. Often, this process takes place in detriment 
of other institutions, such as those in charge of dealing with the adverse effects 
of extraction (Arsel et al. 2016; Bebbington et al. 2018, and for a complemen-
tary view see Dargent, et al. 2017). We argue that scholars should study the 
interactions between the macro-political system and sectoral subsystems where 
the strategies seeking to promote and oppose policy change are deployed. 
Through this perspective we seek to expand current understanding of the influ-
ence of political institutions on policy development. This focus is particularly 
important in contexts where populist politics control governmental initiatives 
of policy reform because of the adverse effects those can have on the quality of 
the policy process in the long-run.
Such an exploration can also contribute to existing discussions about the 
role of populism within democratic systems (Mény & Surel, 2002). These 
studies argue that populism is a political phenomenon that reveal the fail-
ures of democratic systems. However, these failures often refer to system-wide 
institutions of political dynamics, leaving the specifics of subsystem dynamics 
understudied. Knowledge about macro-system and subsystem dynamics will 
improve our understanding of the causal mechanisms tying populism with the 
development of democratic institutions. In particular, a focus on the influence 
of populism in policymaking will shed light on the specific mechanisms by 
which populism reinforces the mechanisms that contain the influence of the 
demos in policymaking.
In this paper, we explore the influence of populism on policymaking by 
studying the role of policy narratives in shaping the process and outcomes of 
82 Revista de Ciencia Política y Gobierno, 5(10), 2018
reforms in the mining sector in Ecuador between 2006 and 2010. The paper 
is organized in 3 sections. In section 1, we present the theoretical background 
of the research. In section 2, we discuss how discourse network analysis can 
contribute to the evaluation of the structure and change of policy networks. 
Section 3 presents the results and the discussion of the study. We close with 
some conclusions.
1. Networks, policy narratives, and policy change 
Three interwoven approaches to the study of policy change populate the lit-
erature on mining governance in Latin America. First, there is an international 
political economy approach that focuses on the outcomes of policy reforms 
seeking to answer the question: who gets what? This approach shows that cleav-
ages between political and economic elites allow the expansion of the extractive 
sector due to their capacity to maintain existing institutions in place. When 
cleavages between elites emerge, challenges to existing policies can form and 
may create conditions for policy change (Bebbington, et al. 2018).
Second, the conflict-centered approach is, predominantly, framed within 
the literature on political ecology (see Bebbington and Bury, 2013; Delgado 
Ramos, 2010; Walter and Martinez-Alier, 2010). The study of conflict shows 
how organized communities and their support networks politicize existing con-
ceptions of the territory or nature to challenge the development of particular 
mining projects. This approach employs a single case-study design that rarely 
offers a replicable analysis of conflicts and their policy outcomes because these 
studies use networks as a metaphor for collective action, leaving the properties 
of these networks understudied.
Finally, there is a post-structuralist approach centered on social and envi-
ronmental justice and identity-politics that investigates parts of the policy pro-
cess -mainly the agenda-setting phase- but is not in dialogue with the literature 
on the policy process. As with the conflict-centered approach, post-structur-
alist studies can also at times be methodologically opaque (see Svampa and 
Antonelli, 2009).
In order to overcome some of the limitations of these studies, we draw on the 
Advocacy Coalition Framework (ACF), one of the leading frameworks in the 
field of policy studies (Nowlin, 2011; Petridou, 2014; Schlager & Blomquist, 
1996). From its inception, the ACF is concerned with the production of 
research that is theoretically grounded, empirically-based, but also transparent 
and reproducible (Sabatier, 1991). In comparison to other frameworks, the 
 Paúl Cisneros / The reform of the Ecuadorian mining sector under Rafael Correa between 2007 and 2010 83
ACF provides a more structured and somewhat parsimonious account of the 
policy process (see Capano and Howlett, 2009). The ACF has two main com-
ponents: the policy subsystem and a set of exogenous variables that describe the 
macro-political system. Competing advocacy coalitions, governmental author-
ities and institutions, and policy outputs and outcomes compose the policy 
subsystem1. The macro-political system includes the so-called stable param-
eters, such as constitutional rules, the distribution of natural resources and 
cultural values, and dynamic parameters, such as changes in socio-economic 
conditions, systemic governing coalitions, public opinion, and decisions made 
in other subsystems (Sabatier, 1988).
When the more stable elements change, long term opportunity struc-
tures open for coalitions to pursue policy change (Jenkins-Smith et al., 2014; 
Sabatier, 1988). Long-term opportunities relate to the openness of the political 
system or the degree of consensus needed for major policy change. Additionally, 
when the relatively dynamic elements change, short-term constraints or addi-
tional resources become available for subsystem actors. Existing ACF literature 
has not explored the role of populist leadership, a political phenomenon of 
increasing interests for scholars in many regions of the world, in creating short 
and long-term opportunities for other policy actors. We take steps towards 
overcoming this gap in the literature by investigating how a populist leader 
influenced the capacity of other actors to participate in the reform of the 
mining sector in Ecuador. 
In the ACF, the critical factor that allows coalitions to change policies is the 
ability that actors have to expand policy networks to mobilize resources and 
make the most out of the emerging opportunities (Ingold, 2011). Coalitions 
will seek change in policies that are compatible with their shared system of 
beliefs. These systems have three layers. First, a layer of core-beliefs that define 
the normative values and the ontological axioms that guide the behavior of 
individuals across policy issues. Second, the policy-core or policy-preferences 
layer that is bounded by the scope of the subsystem. These preferences can be 
normative or empirical, meaning they identify how policies should work or 
the outcomes that specific mixes of policy instruments produce. The beliefs in 
this layer reflect the basic orientation and value priorities or the assessments 
of the seriousness of the problem and preferred solutions for addressing it. 
1 Coalitions include, among others, governmental authorities, activists, scientists and members of the 
press who show a degree of non-trivial coordination and, to an extent, share similar beliefs about policy 
(Sabatier, 1988).
84 Revista de Ciencia Política y Gobierno, 5(10), 2018
Finally, the layer of secondary beliefs includes the means (instruments or tools) 
to achieve the desired policy outcomes (Jenkins-Smith, Nohrstedt, Weible, & 
Ingold, 2017).
When coalitions perceive the opening of an opportunity for change, they 
use their belief system to produce new policy images or interpretations about 
the problems that policies are meant to fix. Baumgartner et al. (2014: 161) 
define a policy image as a «mixture of empirical information and emotive 
appeal» that structures interaction within the subsystem but also connects the 
subsystem to the macro-system.
Policy images are contained in larger narratives that delineate the universe 
of potential forms to understand the substantive issues of a policy problem 
(Shanahan, McBeth, & Hathaway, 2011b). Current scholarship on policy nar-
ratives highlights three strategies used by coalitions to achieve policy change: 
scope of the conflict, causal mechanisms, and devil-angel shifts (Jones & 
McBeth, 2010; Shanahan, Jones, & McBeth, 2011a; Shanahan, McBeth, & 
Hathaway, 2011b).
The producers of policy narratives change the scope of policy conflicts 
depending on the relative position to the outcome. If they are on the losing side, 
they aim to expand the scope of the conflict. Conversely, when policy narra-
tives producers are on the winning side, they engage in narrative strategies that 
contain an issue to the status quo. Given that populist leaders position them-
selves as representatives of ´the people´ that confront the dominant elites and 
institutions (see Levitsky & Loxton, 2013), we could expect to find evidence 
of support for those proposing alternative policy images in early moments of 
the policy process. However, populists are bound by existing democratic insti-
tutions that they seek to control, not change or eliminate; therefore, in later 
stages of the policy process, they defend the institutions from further attacks 
(Mény & Surel, 2002).
When policy actors use causal mechanisms, they arrange narrative elements 
strategically to assign responsibility and blame for a policy problem. There are 
at least four types of causal mechanisms: intentional, inadvertent, accidental, 
and mechanical. An intentional mechanism portrays an actor as purposefully 
causing harm or small benefit to others. The mechanical mechanism denotes an 
arrangement where actions are unguided, but harmful results emerge naturally. 
The inadvertent mechanism depicts a story where some act is producing unfore-
seen negative consequences. Finally, the accidental mechanism is a situation 
where there was no intended outcome guided by directed behavior (Shanahan, 
Jones, McBeth, & Lane, 2013). Regarding this mechanism,  populist leaders 
 Paúl Cisneros / The reform of the Ecuadorian mining sector under Rafael Correa between 2007 and 2010 85
are expected to charge against democratic institutions, in particular, those that 
serve to represent popular interests. Populists argue that representation serves 
only the elites; therefore, political and policy elites, such as political parties and 
the bureaucracy, will be assigned the blame for the shortcomings of a policy 
when populists seek policy change (see Mény & Surel, 2002).
Finally, the angel-devil shifts is the mechanism by which a policy actor iden-
tifies other actors as villains or problem instigators in comparison to how much 
that same actor identifies him- or herself or others as problem fixers or heroes 
(McBeth, Jones, & Shanahan, 2007; Shanahan, Jones, McBeth, & Radaelli, 
2017). A populist leader will create or modify institutions that allegedly pro-
tect the interests of the people because they do not reproduce the shortcomings 
of representative democracy since they are controlled by the leaders’ party or 
movement (Mény & Surel, 2002).
2. Method
Social network analysis provides techniques to analyze the structure and devel-
opment of networks using relational data, which is connections (ties) between 
actors (nodes) considered as non-independent units (Wasserman & Faust, 
1994). In this paper, we study the Ecuadorian mining subsystem between 
2006 and 2010 using a discourse network as a proxy. A discourse network is 
elicited or constructed considering a group of actors and the messages or utter-
ances they make during a specific period. In previous studies of policy change, 
scholars have elicited discourse networks from transcripts of parliamentary 
hearings and newspaper articles convincingly avoiding problems of retrospec-
tive reporting (see Leifeld, 2013, and references therein). We used newspaper 
articles to elicit the discourse network of interest and complemented that infor-
mation with in-depth interviews.
For the mathematical analysis, we broke the discourse network in two. 
A co-occurrence network of actors that show connections between actors if 
they utter the same idea. On the other hand, a co-occurrence network that 
shows connections between utterances if they are emitted by the same actor. 
For example, if two actors express support to the idea that conflicts in the 
mining sector are caused by the mining companies, then a tie or a connection 
is computed between these actors. In the second network, an utterance has a 
connection to another if an actor emits both during the period under analysis.
86 Revista de Ciencia Política y Gobierno, 5(10), 2018
Using computational techniques, we measured changes in the structure of 
the discourse network considering two variables. The first one measures the 
relative position of nodes with respect to others in a network. This measure, 
known as centrality, provides an assessment of which actors are key (more cen-
tral) in conveying messages to others in the network. We also study density or 
the number of connections present within a group of nodes in the network. 
Higher density suggests more communication between policy actors, or more 
compatibility between utterances, during the policy process. These variables 
should change as the policy process unfolds to show a prominent role of the 
populist leader in policymaking. 
We coded the type of actors inductively, and for the utterances, we used 
the definition of the components of belief systems proposed by Sabatier 
and Jenkins-Smith (1999). To analyze the components of policy narratives, 
we coded causal mechanisms using binary variables (present or not present). 
Finally, the angel and devil shifts were identified, considering the changes in 
the number of connections between the populist leader and the pro-change 
and the pro-status-quo coalitions considering four sub-periods. T1 is the year 
before the election of Rafael Correa as president, T2 corresponds to the rise of 
Rafael Correa to power in 2007, T3 includes the Constitutional Assembly of 
2008 and the legislative process to write a new sectoral law, T4 corresponds to 
the first year of implementation of the law2.
The core of our dataset consists of 816 online news reports from news 
sources published in Ecuador between 2006 and 20103. From these, we 
extracted explicit quotations of utterances that were deductively coded using 
Discourse Network Analyzer (DNA), a category-based content analysis 
2 In this paper, we do not elaborate on the interaction of discourse about the mining sector with another 
contextual variable such as the macroeconomic situation for two reasons. First, during the period of 
analysis, mineral commodity prices (especially gold and copper) rose in a sustained manner; therefore, the 
government did not have to adapt its discourse to changing market conditions. Second, the government 
maintained high levels of public spending on social programs and infrastructure development during this 
period, contracting debt even when the prices of oil decreased in 2009. To cover the budgetary deficit, 
the government resorted to pre-selling oil (see Caria & Martín, 2009; Ruiz & Iturralde, 2013), which 
maintained the pressure to exploit the country´s metallic endowments.
3 The sources for articles are the websites Explored, El Comercio, and EcuadorInmediato.com. They cover 
all major newspapers in the country for the entire period under analysis. The keywords (in Spanish) 
minería (mining) and mina (mine) were used to retrieve the articles. The period for analysis was defined 
considering two aspects. First, the number of reports before 2006 going back to 1994 was minimal; we 
found less than five reports per year. Second, the Legislative Commission approved the new mining law 
in 2009. Additionally, in 2010, the dispute between Correa and the privately-owned media changed the 
coverage of events with media owned by the state reporting for the government and the private media in 
stark opposition.
 Paúl Cisneros / The reform of the Ecuadorian mining sector under Rafael Correa between 2007 and 2010 87
s oftware developed by Leifeld (2013). All graphical representations and net-
work measurements were generated using Gephi 0.9.2. One researcher coded 
all the newspaper articles in three waves. All articles in each wave were coded 
and then coded again after two weeks. Stability was the (weak) criteria used to 
control the quality of the coding process (Krippendorff, 2004). Intra-observer 
inconsistencies were roughly 2% across waves.
Communicative validation, that is, a consensus achieved discursively about 
the content of the analysis between researchers and the researched (Kohlbacher, 
2006; Mayring, 2014), was used to assess the validity of the coding phase. 
We used in-depth semi-structured interviews with coalition elites from pro-
change (n=5) and pro-status-quo organizations (n=4) to validate the timeline 
of events leading to network change and to identify the main actors and the 
main issues of contention. Respondents suggested minor changes in the struc-
ture of the networks — most of these changes related to the names of some 
civil society organizations.
3. Results and discussion
3.1. Central issues in the mining subsystem before the reform
During the decade of the 1990s, Ecuador reformed its mining code to attract 
foreign capital, and as many other countries in Latin America, started envi-
ronmental standards modernization (Zimmerer & Carter, 2002). In 1991, 
the National Congress passed a new mining law that, among others, elimi-
nated State-owned mining companies, introduced tax incentives for foreign 
investments, and lowered royalty payments (Cisneros, 2011). After a brief 
period of growth, the mining sector was paralyzed due to the financial crisis 
in Southeast Asia in 1996. To attract more investments, in 2001, the govern-
ment of Ricardo Noboa introduced reforms to the mining law, increasing fiscal 
incentives. Petitions of new areas for exploration grew so fast that, by 2003, the 
sectoral authority had dropped all oversight of existing operations and devoted 
its resources solely to granting exploration permits (Liebenthal et al., 2003).
On the environmental and social fronts, Ecuador ratified Convention 169 
of the International Labor Organization on Free, Prior and Informed Consent 
to Indigenous and Tribal Peoples in 1998, which mandates that governments 
seek consent from indigenous leaders before implementing projects that may 
affect local livelihoods. Further consultations to local stakeholders potentially 
affected by mining operations were introduced in the constitutional reform 
88 Revista de Ciencia Política y Gobierno, 5(10), 2018
of 1998 and regulated in the 1999 General Environmental Law as part of pre-
liminary environmental assessments.
In the remainder of this section, we show how the discourse network related 
to the mining sector in Ecuador evolved from this point onwards. We analyze 
the changes in the composition of the network in terms of the relative position 
of policy actors and the elements of their belief systems across the four sub-
periods of interest.
3.2. Policy change and the mining policy subsystem
In 2006, the dominant policy image of the mining sector entered in crisis. 
The dispute about the future of the mining sector started during the admin-
istration of Alfredo Palacio (2004-2006). In this period, anti-mining orga-
nizations from different regions of the country clashed with security forces, 
both public and private, that protected concessions during protests against the 
development of large-scale mines (Cisneros, 2011).
The analysis of media reports shows that organized anti-mining groups dis-
puted the core idea that large-scale mining promotes sustainable development. 
Based on the experience of conflictive relations with mining companies over 
the 1990s, they argued that large-scale mining undermines all forms of capi-
tals: social, environmental and even economic. These rivaling interpretations 
of the effects of large-scale mining were the most frequent elements of the dis-
course network before Correa took office in 2007. They constitute antagonistic 
definitions of the ontology of large-scale mining, which speaks of a highly 
polarized policy subsystem.
Table 1 shows the most central utterances made by the policy actors of the 
mining policy subsystem in 2006. As shown in the Degree column of Table 1, 
these rivaling images were uttered by the same number of actors. However, the 
critical policy image had a more central role in the discourse network (higher 
score of betweenness centrality) because it was connected to a higher number 
of other highly central utterances. This structure suggests that critics of the 
existing policy image were using a diversity of discursive elements intertwined 
in a complex narrative to politicize the effects of large-scale mining.
Figure 1 shows the structure of the discourse network in 2006 elicited from 
media reports. The nodes in the network represent utterances, and their size 
corresponds to the normalized scores of betweenness centrality. The right-hand 
side of the network (in green) corresponds to the pro-change coalition while 
the left-hand side of the network corresponds to the pro-status-quo coalition 
(in red). Higher betweenness centrality suggests that larger nodes sit between 
 Paúl Cisneros / The reform of the Ecuadorian mining sector under Rafael Correa between 2007 and 2010 89
the paths that connect any other two nodes more often than others. The most 
central utterance (in yellow), is a secondary belief used by pro-change actors to 
demand a halt to new concessions, and audits for existing concessions for their 
future return to the State. This means that actors on rivaling sides (all nodes 
in depicted green) of the network supported the opening of negotiations to 
reform the existing policy, but this required taking stock of the existing effects 
of large-scale operations. 
Table 1 also shows the type of utterance according to its position in the 
belief system of the actors that employ it. The distribution of the types of utter-
ances is shown in more detail in Table 2. 
Table 1. Betweenness centrality and popularity of utterances in 2006
Betweenness 
Utterance Policy belief Degree
Centrality
Halt new concessions, audit & revert unlawful concessions 0.169548 Secondary 13
Conflicts emerge from companies and government 0.167135 Policy preference, 9
targeting dissent empirical
Large Scale Mining undermines all forms of capital 0.164849 Core 10
Large Scale Mining promotes development 0.120833 Core 10
Form ad hoc commissions to audit existing concessions 0.083611 Secondary 10
Conflicts emerge from electoral interest or ideological 0.040833 Secondary, origin 9
extremism of conflicts
The 2001 neoliberal law is detrimental to the interests of 0.031389 Policy preference, 9
the State empirical
Reform the 2001 law to increase revenue for the State 0.031389 Policy preference, 9
normative
Conflicts emerge from lack of free prior and informed 0.030468 Policy preference, 6
consent (FPIC) empirical
Table 2 shows the percentage of each type of utterance by type of actor. 
In both coalitions, most actors uttered policy preferences more than they did 
policy core beliefs and secondary beliefs. Additionally, empirical policy prefer-
ences were the most common in both coalitions. In T1, actors disputed inter-
pretations of the effects of the existing law and the causes of conflicts more 
than they did ontological and axiological elements of their belief systems. This 
finding suggests that despite the polarized structure of two coalitions bridged 
by just one actor, the subsystem was relatively open to actors and ideas and 
that consensus about the future of large-scale mining could have been pursued 
between them.
90 Revista de Ciencia Política y Gobierno, 5(10), 2018
Table 2. The composition of discourse network in 2006
Percentage Sectoral. Political Mining Local National Company-
Belief NGO CSO
total Authority Party guild govt govt. Contractor
Core 21% 17% 26% 0% 0% 0% 14% 0% 0%
Policy 50% 39% 28% 40% 50% 50% 43% 0% 0%
preference
Normative 29% 9% 5% 30% 17% 0% 0% 0% 0%
Empirical 21% 30% 23% 10% 33% 50% 43% 0% 0%
Secondary 29% 4% 19% 20% 0% 0% 0% 100% 0%
Core 20% 0% 33% 25% 0% 50% 0% 0% 33%
Policy 60% 0% 33% 25% 0% 25% 0% 50% 0%
preference
Normative 0% 0% 0% 0% 0% 0% 0% 0% 0%
Empirical 60% 0% 33% 25% 0% 25% 0% 50% 67%
Secondary 20% 0% 0% 25% 0% 0% 0% 0% 0%
The network of actors has a similar structure to the network of utterances 
depicted in Figure 1. The two rivaling coalitions are connected by just one 
node, in this case, the Ministry of Energy and Mines (MEM). Table 3 shows 
the centrality score for all actors and identifies the type of actors present. 
 Pro-change actors had, on average, higher centrality scores than pro-status quo 
actors, which speaks of a larger and more connected coalition. Additionally, 
from the information contained in Table 3, we can determine that during 2006 
roughly 80% of actors critical of large-scale mining were civil society organiza-
tions and non-governmental organizations. These organizations included local 
grass-roots organizations, as well as regional and national coalitions. On the 
other hand, roughly 60% of the actors that promoted a positive image of large-
scale mining were industry actors such as mining companies, and the chamber 
of mining.
pro-status-quo pro-change
Figure 1. Discourse network in 2006: utterances
92 Revista de Ciencia Política y Gobierno, 5(10), 2018
Table 3. Actors in the discourse network during 2006
Actor Type Betweenness centrality
Ministry of Energy and Mines (MEM) Sectoral Authority 0.192421
PACHAKUTIK Political Party 0.157958
Foro de los Pueblos Afectados por la Mineria CSO 0.133213
Consejo de Comunidades de Intag CSO 0.101414
Acción Ecológica (AE) NGO 0.091406
Comite Defensa Vida Morona CSO 0.081371
DECOIN NGO 0.081154
Ascendant Copper Mining company 0.039211
Coordinadora Zonal de Intag CSO 0.03178
Municipality of Cotacachi Local govt 0.014223
Consejo de Comunidades de Garcia Moreno CSO 0.014134
Ingenieros geólogos Mining guild 0.012587
Junta Parroquial de Peñaherrera Local govt 0.006885
UNAGUA CSO 0.005741
Asamblea Nacional Ambiental (ANA) CSO 0.001635
Cámara de Minería del Ecuador (CME) Mining guild 0.00135
Organizacion Desarrollo Intag Contractor 0.00135
Canton Naranjal CSO 0.000854
UNORCAC CSO 0.000273
The remainder of this section discusses the evolution of the discourse net-
work highlighting that a) at all times, the network included one pro-change 
and one pro-status-quo coalition; b) both challengers and dominant actors 
expanded the conflict in the early stages of the policy cycle; c) as a populist 
leader seeking to change the sectoral policy, Rafael Correa became the bridge 
between these rivaling coalitions; then d) he used angel and devil-shifts to 
reduce existing institutional friction, and to e) position himself as the domi-
nant actor in the discourse network.
a) A polarized mining subsystem
Figure 2 shows the structure of the discourse network of actors in the four 
sub-periods under analysis. In all cases, the red nodes correspond to the pro-
status-quo coalition, and the green nodes represent the pro-change coalition. 
 Paúl Cisneros / The reform of the Ecuadorian mining sector under Rafael Correa between 2007 and 2010 93
The most central actor in the network is depicted in blue, and the size of the 
nodes corresponds to their normalized betweenness centrality.
Figure 2. Co-occurrence network of actors
Two elements are critical for the comparison of the four periods. First, the 
pro-change coalition always had more actors than the pro-status-quo coalition. 
Table 4 presents the count of actors and the density of each coalition for each 
period. The difference in size and density suggests more complexity in the pro-
change coalition.
Table 4. Evolution of the network of actors
T1 T2 T3 T4
2006-2007 2007-2008 2008-2009 2009-2010
Coalitions Density Nodes Density Nodes Density Nodes Density Nodes
Pro-change 0.399 32 0.504 38 0.643 19 0.382 11
Pro-status-quo 0.857 8 0.691 17 0.273 11 0.476 7
94 Revista de Ciencia Política y Gobierno, 5(10), 2018
Second, the analysis of the density of the connections between actors in 
each coalition shows that the pro-status-quo coalition was relatively denser 
during the agenda-setting phase (T2) compared to the decision-making (T3) 
and implementation (T4) phases. The most radical members of this coalition 
lost connection during T2 when Rafael Correa took office using a nationalistic 
rhetoric that foreign private companies and some of their national counter-
parts opposed, as depicted in the left lower corner of Figure 2. Reports from 
the media show that Correa confronted the mining companies over the idea 
that the existing law was detrimental to the interests of the State. The same 
source shows that some pro-status-quo actors expressed deep opposition to the 
principles of the reform, arguing that under the existent regulations, large-scale 
mining was effectively contributing to sustainable development. 
b) Conflict expansion and containment
Figure 3 shows that during agenda-setting (between 2006 and 2008), the total 
number of actors in the pro-change coalition increased by roughly 30%, while 
the total count of appearances of actors in the discourse network increased by a 
factor of three. On the other hand, the number of actors in the pro-status-quo 
coalition doubled, while their frequency of appearance nearly quadrupled.
Figure 3. Number of policy actors per coalition
pro-change pro-status-quo
40 400 8 250
200
30 300 6
150
20 200 4
100
10 100 2
50
0 0 0 0
2006-07 2007-08 2008-09 2009-10 2006-07 2007-08 2008-09 2009-10
CSO Sect. Authority OSC Sectoral Authority
NGO Local govt.
Mining guild Companies-contractor
Mining guild Experts
Political party Central govt. Central government Experts
Companies-contractor Frequency count Political party Frequency count
 Paúl Cisneros / The reform of the Ecuadorian mining sector under Rafael Correa between 2007 and 2010 95
In the same period, the number of utterances favorable to change increased 
only slightly (see Figure 4). However, their frequency quadrupled. The utter-
ances that increased the most, correspond to core beliefs and policy preferences. 
Meanwhile, the number of utterances defending the status-quo increased mar-
ginally for both core beliefs and policy preferences. The frequency of these 
utterances grew by a factor of five, in particular, the utterances regarding policy 
preferences.
Figure 4. Utterances of policy beliefs per coalition
pro-change pro-status-quo
8 250 5 120
7
200 4 100
6
5 80
150 3
4 60
3 100 2
40
2 50 1
1 20
0 0 0 0
2006-07 2007-08 2008-09 2009-10 2006-07 2007-08 2008-09 2009-10
Core beliefs Policy preferences Core beliefs Policy preferences
Secondary beliefs Total count Secondary beliefs Total count
These findings suggest that the two coalitions engaged in network expan-
sion, both in terms of the number of supporters that joined the discourse net-
work and the frequency in which they appeared. Although the increase in the 
frequency of appearance could be explained by an increasing interest of the 
media in covering this salient policy issue, the analysis of the content of the 
utterances shows that the rise of a populist leader to power created uncertainty 
for both challengers and incumbents. Hence, both coalitions may have sought 
expansion to increase their resources for political struggle. This finding adds 
nuance to the idea that the revisionist coalition will expand conflict while the 
incumbents will contain the conflict as it is presented in the Policy Narratives 
Framework (see Shanahan, Jones, & McBeth, 2011a, p. 543). The presence 
of a populist leader changes this dichotomous view of incumbents and chal-
lengers, adding a third actor that could sway policy narratives towards both 
sides demanding conflict expansion from all the involved actors.
Both coalitions relied on core beliefs and policy preferences to expand the 
discourse network during T2 or the agenda-setting phase. Data shows that the 
number of secondary utterances diminished slightly for both. Table 5 shows 
96 Revista de Ciencia Política y Gobierno, 5(10), 2018
that while the number of normative policy preferences increased significantly, 
the number of empirical preferences dropped in for both coalitions. These 
changes speak of polarization of the political struggle within the policy sub-
system as the policy process advanced.
Table 5. Distribution of policy preferences by coalition
T1 T2
Coalition Policy preference
2006-2007 2007-2008
Normative 8% 37%
Pro-change
Empirical 45% 18%
Normative 0% 20%
Pro-status-quo
Empirical 56% 40%
Scholars of the Policy Narratives Framework (PNF) argue that not all coali-
tions will be as effective in using policy narratives strategically. Advocacy coali-
tions with policy narratives that contain higher levels of coalitional glue will 
more likely influence policy outcomes (Shanahan, Jones, McBeth & Radaelli, 
2017, p. 191). Our data partially confirms this idea. As Figure 2 showed, the 
pro-change coalition incorporated a larger diversity of actors that uttered more 
diverse elements of their belief systems, configuring a complex policy narra-
tive in all periods. This coalition was always less dense than the pro-status-quo 
rivals; therefore, it had less coalitional glue. As predicted by the PNF, the pro-
change coalition did not gain control of the subsystem during implementation. 
Its diversity only allowed for some of the elements of their belief system to be 
incorporated by other actors in the new law because they were instrumental in 
opening an opportunity for change. However, the more cohesive network did 
not maintain its dominant position at the end of the policy cycle. As we discuss 
in the next subsection, the angel and devil shifts used by the populist leader to 
deal with existing institutional friction presented by this coalition was vital to 
produce a reform that combined elements of the pro-change and the pro-status 
quo coalitions.
c) The populist leader as a catalyst for change
The crisis in the mining sector was a relatively salient issue in the mainstream 
media during 2006, and Rafael Correa became a prominent actor in the dis-
course network after his inauguration as president in January 2007. Moreover, 
President Correa became the main connector of the pro-change and the 
 pro-status-quo coalitions. Despite the appointment of a pro-change minister 
 Paúl Cisneros / The reform of the Ecuadorian mining sector under Rafael Correa between 2007 and 2010 97
in the MEM in 2007, Correa replaced MEM as the most central actor in the 
discourse network.
Table 6 shows the betweenness centrality for selected actors between 2006 
and 2010. It shows that president Correa became more central to the net-
work as the policy-cycle evolved. Correa maintained a dominant position in 
the network well above the sectoral authority MEM even after the new law 
was approved. This finding suggests that the personalistic approach to politics 
characteristic of populist leaders as highlighted in the literature is transferred to 
the realm of policymaking, affecting the evolution of sub-systemic dynamics.
Table 6. Normalized betweenness centralities
Betweenness centrality
T1 T2 T3 T4
Actor Type
2006-2007 2007-2008 2008-2009 2009-2010
Rafael Correa President N/A 0.28 0.29 0.39
MEM Sectoral Authority 0.27 0.13 0.00 0.03
CNVDS CSO N/A 0.10 0.01 N/A
CME Mining guild 0.00 0.06 0.00 0.06
UNAGUA CSO 0.00 0.01 0.00 0.00
PACHAKUTIK Political Party 0.11 0.01 0.05 0.17
MPD Political Party N/A 0.01 0.00 N/A
DECOIN CSO 0.09 0.00 N/A N/A
In terms of utterances, or the components of the belief system deployed in 
the policy process, those defended by President Correa were always more cen-
tral in the network; however, not always the most frequent. Correa’s utterances 
became the most frequent in the last period (2009-2010) after the new law 
was approved, and implementation started. This finding suggests that compat-
ibility of the contents of policy beliefs is more important than the frequency of 
utterance to position an actor as dominant in a discourse network.
d) Populist politics: angle and devil-shifts
Correa used a combination of strategic appointments and a plebiscite to reduce 
the existing institutional friction in the mining sector and skew the system 
against his opponents. Initially, the dominant actors of the network opposed 
the reforms arguing that the existing law allowed the mining industry to pro-
mote sustainable development and that the conflicts in the mining sector origi-
nated in electoral interests and ideological extremism. Pro-status-quo interests 
98 Revista de Ciencia Política y Gobierno, 5(10), 2018
effectively mobilized to block the proposed reform in Congress. To break this 
policy monopoly, Correa appointed a supporter of reform as minister of mines 
and then convened a Constituent Assembly where the new principles for the 
policy image were set. During this period (2007-2009), Rafael Correa, the 
most central actor in the network, uttered the messages depicted in Figure 5. 
Utterances are represented as red nodes, and they are scaled considering the 
number of mentions they received during the period. They are ordered clock-
wise considering the dates when Correa uttered them.
From right to left, Correa set the tone for the upcoming reform; he recog-
nized the importance of the mining sector and proposed to create a «special-
ized ministry» (Rafael Correa, El Comercio, 21-2-2007). Then he criticized 
the existing configuration of the mining subsystem labeling it as a very «messy 
sector» (Rafael Correa, El Comercio, 5-06-2007) and proposed to order audits 
to check if «multinational companies are affecting water catchments» (Rafael 
Correa, El Comercio, 6-30-2007). Up to this point, Correa used an inten-
tional causal mechanism constructing a negative image of the subsystem and 
its dominant actors. For this narrative, he adopted some elements introduced 
in the discourse network by pro-change actors in the previous period, such as 
the discussion of the effects of mining over water catchments.
Later, Correa criticized those same actors which resumed mobilizations 
against the government’s plan to develop the mining sector. Again, he used 
an intentional causal mechanism to label anti-mining activists as «irrespon-
sible people who want to live sitting above such richness, who want to leave 
it in the ground due to their fundamentalism» (Rafael Correa, El Comercio, 
8-11-2008). At the same time, Correa adopted the idea of mining as an industry 
that uses environmentally sound practices, and along with the mining guild 
spoke of «responsible mining» (Rafael Correa, El Comercio, 21-10-2019).
Figure 5. Utterances made by Rafael Correa during the reform
 Paúl Cisneros / The reform of the Ecuadorian mining sector under Rafael Correa between 2007 and 2010 99
These findings show that angel and devil shifts served an actor with authori-
tative power in a polarized subsystem to reduce the institutional friction and 
accomplish policy change. Narratives, however, are not sufficient conditions 
for achieving this result. In this case, the mobilization of legal power and voter 
support to the Constitution proposed by Correa´s political movement served 
to break the existing policy monopoly.
e) The consolidation of the populist leader
Current literature on Latin America’s populist leaders of the ‘Pink Tide’ high-
light the personalistic approach to politics used by these actors in pursuing state 
reforms. Our study shows that the realm of policy reforms also exhibits this char-
acteristic. Our data show a change in the composition of the discourse network 
of the mining policy subsystem that suggest a loss of diversity and openness.
Figure 4 shows a decaying number of actors between 2008 and 2010 when 
the legislative process unfolded in the Constitutional Assembly and then in the 
Legislative Commission that approved the new mining law. As Howlett et al. 
(2009) argue, this dynamic is expected as the policy cycle unfolds because of 
growing specialization in the discussion of policy alternatives and instruments. 
However, it is noticeable that the number of civil society organizations, unlike 
any other type of actors, diminished more dramatically.
Civil society organizations (CSO) had the highest numbers, in particular, 
during T2 or the 2007-2008 period. By 2010, the number of CSOs fell below 
the initial numbers seen in 2006. The second most popular type of actor, NGOs 
also dropped between 2006 and 2010. Additionally, Table 6 shows declining 
centralities for the most central actor of the pro-change coalition, the umbrella 
organization CNDVS formed in 2007. CNDVS’s centrality declined as the 
policy-cycle evolved until it disappeared from the network in 2009. In this 
year, CNDVS´s organizations resumed attempts to challenge the new policy 
independent of each other; therefore, we see a slight increase in the centrali-
ties of the regional organization UNAGUA, and the indigenous political party 
PACHAKUTIK.
Regarding the frequency of appearance of each type of actor, the sectoral 
authority MEM rose to prominence during T2 or the 2007-2008 period 
when along with Rafael Correa it mediated between the two sides of the net-
work. However, unlike Correa’s frequency of appearance, that of the sectoral 
authority fell between 2008 and 2010. This difference indicates that after the 
reform, subsystem interactions remained under the control of a macro-system 
actor and not the sectoral authority as is expected.
100 Revista de Ciencia Política y Gobierno, 5(10), 2018
As we mentioned above, Correa’s utterances became the most central of 
the discourse network as the policy process evolved and they also became the 
most frequent. In the 2009-2010 period, the most frequent utterance made by 
Correa was about the environmental soundness of the new policy because the 
institutions created by the Constitution had clear mandates oriented towards 
guaranteeing the right of the people and nature. He highlighted that after the 
reform, the State had the means to effectively control mining operations to 
promote sustainable development in the mining regions. 
Figure 3 shows that the number of utterances that refer to core beliefs 
increased between 2006 and 2010, while policy preferences and secondary 
beliefs declined in the same period. This finding suggests a growing polariza-
tion of narratives during the reform because core beliefs are harder to negotiate 
between rivaling coalitions than policy preferences or secondary beliefs (Weible 
& Sabatier, 2007a).
As we showed above, the devil shift performed by Correa during the legisla-
tive process explains the reduction in the number of actors from organized civil 
society. There are strong indications that the judicialization of protest during 
this period played a significant role in de-mobilizing organized civil society 
(Cárdenas, Jaramillo, & Loachamín, 2011). In sum, the findings in this section 
suggest that as a result of the reform, the policy subsystem became less diverse 
and less open to actors and ideas.
Conclusion
Populist leaders can undermine democratic institutions and politics, but the 
mechanisms by which they influence policymaking processes do not often 
receive enough attention from scholars. In this paper, we have shown that 
changes in the macro-political system, where a populist leader deployed anti-
establishment rhetoric, influenced reforms to the mining sector, reducing the 
openness and diversity of a policy subsystem. In this way, populist strategies 
have undermined policymaking by constraining the influence of civil society 
organizations in it. This outcome will likely deepen the limited capacity that 
public sector organizations, as well as mining companies, have to manage the 
adverse effects of mining on local populations and the environment.
The Advocacy Coalition Framework provided essential theoretical tools to 
analyze the interaction between changes in the macro-political system and the 
policy subsystem, mainly through the concept of the belief system. In combina-
tion with discourse analysis, the ACF’s lenses on policy change and s ubsystem 
 Paúl Cisneros / The reform of the Ecuadorian mining sector under Rafael Correa between 2007 and 2010 101
dynamics highlight the power of policy narratives as resources deployed by 
populists to increase their influence over the policy process. It shows that pop-
ulists break down existing policy monopolies by strategically using alterna-
tive policy images and their legal capacity to reduce institutional friction and 
promote policy change. This strategy influences the expansion of conflict by 
existing coalitions and not just by the reformists, as current theory proposes. 
Additionally, in order to cement their control over policy subsystems, populist 
leaders use angel and devil shifts producing reforms that combine policy beliefs 
from both reformists and incumbents.
Future studies interested in the interactions between populism and policy-
making should focus on the long-term effects of policy change under popu-
list regimes to expand our understanding of the capacity of policy actors to 
regenerate the diversity and openness of a policy subsystem. The use of larger 
data-sets and more advanced social network analysis techniques would further 
this objective. This approach is still in its infancy for the analysis of policies in 
Latin America, but we hope this article will encourage others to follow this line 
of inquiry.
References
Arsel, M., Hogenboom, B., & Pellegrini, L. (2016). The extractive imperative and the boom in 
environmental conflicts at the end of the progressive cycle in Latin America. The Extractive 
Industries and Society, 3(4), 877-879. https://doi.org/10.1016/j.exis.2016.10.013
Baumgartner, F. R., Jones, B. D., & Mortensen, P. B. (2014). Punctuated Equilibrium Theory: 
Explaining Stability and Change in Public Policymaking. In P. A. Sabatier & C. M. Weible 
(Eds.), Theories of the Policy Process (pp. 25-58). Boulder: Westview Press.
Bebbington, A., & Bury, J. (2013). Subterranean struggles: New dynamics of mining, oil, and gas 
in Latin America. Austin: University of Texas Press.
Bebbington, A., Abdulai, A.-G., Humphreys Bebbington, D., Hinfelaar, M., & Sanborn, C. (2018). 
Governing Extractive Industries. Oxford: Oxford University Press. https://doi.org/10.1093/
oso/9780198820932.001.0001
Capano, G. & Howlett, M. (2009). Introduction: The Determinants of Policy Change: Advancing 
the Debate. Journal of Comparative Policy Analysis: Research and Practice, 11(1), 1-5.
Cárdenas, A., Jaramillo, L., & Loachamín, R. N. (2011). Los escenarios de la criminalization a defen-
sores de Derechos Humanos y de la Naturaleza en Ecuador: Desafios para un Estado Constitucional 
De Derechos. Quito: Defensoría del Pueblo de Ecuador.
Caria, S., & Domínguez Martín, R. (2015). Estructura productiva primario-exportadora y Buen 
Vivir en Ecuador, 2009-2014. Revista Internacional de Cooperación Y Desarrollo, 2(1), 1-28. 
https://doi.org/10.21500/23825014.2232
Cisneros, P. (2011). ¿Cómo se construye la sustentabilidad ambiental? Análisis de experiencias conflic-
tivas de la industria minera en el Ecuador. Quito: FLACSO Ecuador.
102 Revista de Ciencia Política y Gobierno, 5(10), 2018
Dargent, E., Orihuela, J. C., Paredes, M., & Ulfe, M. E. (Eds.). (2017). Resource Booms 
and Institutional Pathways. Chaltenham: Springer International Publishing. https://doi.
org/10.1007/978-3-319-53532-6
Delgado-Ramos, G. C. (Ed.). (2010). Ecología Política de la Minería en América Latina. 
Mexico: UNAM.
Haarstad, H. (2012). New Political Spaces in Latin American Natural Resource Governance. 
New York: Palgrave Macmillan. https://doi.org/10.1057/9781137073723
Howlett, M., Ramesh, M., & Perl, A. (2009). Studying Public Policy. Toronto: Oxford University Press.
Ingold, K. (2011). Network Structures within Policy Processes: Coalitions, Power, and 
Brokerage in Swiss Climate Policy. Policy Studies Journal, 39(3), 435-459. https://doi.
org/10.1111/j.1541-0072.2011.00416.x
Jenkins-Smith, H. C., Nohrstedt, D., Weible, C. M., & Sabatier, P. A. (2014). The Advocacy 
Coalition Framework: Foundations, Evolution, and Ongoing Research. In P. A. Sabatier & C. 
M. Weible (Eds.), Theories of the Policy Process (pp. 183-223). Boulder: Westview Press.
Jenkins-Smith, H. C., Nohrstedt, D., Weible, C. M., & Ingold, K. (2017). The Advocacy 
Coalition Framework: An Overview of the Research Program. In C. M. Weible & P. A. Sabatier 
(Eds.), Theories of the Policy Process (pp. 135-172). Boulder: Westview Press. https://doi.
org/10.4324/9780429494284-5
Jones, M. D., & McBeth, M. K. (2010). A Narrative Policy Framework: Clear Enough to Be Wrong? 
Policy Studies Journal, 38(2), 329-353. https://doi.org/10.1111/j.1541-0072.2010.00364.x
Kohlbacher, F. (2006). The use of qualitative content analysis in case study research. Forum 
Qualitative Sozialforschung/Forum on Qualitative Sociology, 7(1).
Krippendorff, K. H. (2010). Content Analysis: An Introduction to Its Methodology. Thousand 
Oaks: SAGE.
Leifeld, P. (2013). Reconceptualizing Major Policy Change in the Advocacy Coalition Framework: 
A Discourse Network Analysis of German Pension Politics. Policy Studies Journal, 41(1), 
169-198. https://doi.org/10.1111/psj.12007
Levitsky, S., & Loxton, J. (2013). Populism and competitive authoritarianism in the Andes. 
Democratization, 20(1), 107-136. https://doi.org/10.1080/13510347.2013.738864
Liebenthal, A., Dahan, C., Babelon, D., (2003). Project Performance Assessment Report: Ecuador 
Mining Development and Environmental Control Technical Assistance Project (Loan 2655-
EC). Washington DC: World Bank.
Mayring, P. (2014). Qualitative content analysis: theoretical foundation, basicprocedures and soft-
ware solution. Klagenfurt: GESIS. https://doi.org/10.1007/978-94-017-9181-6_13
Mény, Y., & Surel, Y. (2002). The Constitutive Ambiguity of Populism. In Y. Mény & Y. 
Surel (Eds.), Democracies and the Populist ChallengePalgrave. (pp. 1-24). https://doi.
org/10.1057/9781403920072_1
McBeth, M. K., Jones, M. D., & Shanahan, E. A. (2007). The Narrative Policy Framework. 
In C. M. Weible & P. A. Sabatier (Eds.), Theories of the Policy Process (pp. 225-266). Boulder: 
Westview Press.
Nem Singh, J., & Bourgouin, F. (Eds.). (2013). Resource governance and developmental states in the 
Global South: Critical international political economy perspectives. New York: Springer. https://
doi.org/10.1057/9781137286796
Nowlin, M. C. (2011). Theories of the Policy Process: State of the Research and Emerging Trends. 
Policy Studies Journal, 39(s1), 41-60. https://doi.org/10.1111/j.1541-0072.2010.00389_4.x
Petridou, E. (2014). Theories of the Policy Process: Contemporary Scholarship and Future 
Directions. Policy Studies Journal, 42(s1), 12-32. https://doi.org/10.1111/psj.12054
 Paúl Cisneros / The reform of the Ecuadorian mining sector under Rafael Correa between 2007 and 2010 103
Ruiz Acosta, M., & Iturralde, P. (2013). La alquimia de la riqueza. Estado, petróleo y patrón de 
acumulación en Ecuador. Quito: CDES.
Sabatier, P. A. (1988). An advocacy coalition framework of policy change and the role of policy-
oriented learning therein. Policy Sciences, 21(2-3), 129-168. https://doi.org/10.1007/
BF00136406
Sabatier, P. A. (1991). Toward Better Theories of the Policy Process. PS: Political Science & Politics, 
24(02), 147-156. https://doi.org/10.1017/S1049096500050630
Sabatier, P. A., & Jenkins-Smith, H. C. (1999). The Advocacy Coalition Framework: An Assessment. 
In P. A. Sabatier (Ed.), (pp. 117-168). Boulder: Westview Press.
Schlager, E., & Blomquist, W. (1996). A Comparison of Three Emerging Theories of the Policy Process. 
Political Research Quarterly, 49(3), 651-672. https://doi.org/10.1177/106591299604900311
Shanahan, E. A., Jones, M. D., & McBeth, M. K. (2011a). Policy Narratives and Policy Processes. 
Policy Studies Journal, 39(3), 535-561. https://doi.org/10.1111/j.1541-0072.2011.00420.x
Shanahan, E. A., Jones, M. D., McBeth, M. K., & Lane, R. R. (2013). An Angel on the Wind: 
How Heroic Policy Narratives Shape Policy Realities. Policy Studies Journal, 41(3), 453-483. 
https://doi.org/10.1111/psj.12025
Shanahan, E. A., Jones, M. D., McBeth, M. K., & Radaelli, C. M. (2017). The Narrative Policy 
Framework. In C. M. Weible & P. A. Sabatier (Eds.), Theories of the Policy Process (pp. 173-1213). 
Boulder: Westview Press. https://doi.org/10.4324/9780429494284-6
Shanahan, E. A., McBeth, M. K., & Hathaway, P. L. (2011b). Narrative Policy Framework: 
The Influence of Media Policy Narratives on Public Opinion. Politics & Policy, 39(3), 373-400. 
https://doi.org/10.1111/j.1747-1346.2011.00295.x
Svampa, M., & Antonelli, M. A. (Eds.). (2010). Minería transnacional, narrativas del desarrollo y 
resistencia sociales. Buenos Aires: Editorial Biblos.
Walter, M., & Martinez-Alier, J. (2010). How to be heard when nobody wants to listen: com-
munity action against mining in Argentina. Canadian Journal of Development Studies/Revue 
Canadienne D’études Du Développement, 30(1-2), 281-301. https://doi.org/10.1080/02255
189.2010.9669292
Wasserman, S., & Faust, K. (1994). Social Network Analysis. Cambridge: Cambridge University 
Press. https://doi.org/10.1017/CBO9780511815478
Weible, C., & Sabatier, P. (2007). A Guide to the Advocacy Coalition Framework. In Handbook of 
public policy analysis: theory, politics, and methods (pp. 123-136). Boca Ratón: CRC/Taylor & 
Francis. https://doi.org/10.1201/9781420017007.pt3
Zimmerer, K., & Carter, E. (2002). Conservation and Sustainability in Latin America and the 
Caribbean. Yearbook. Conference of Latin Americanist Geographers, 27, 207-249.