Ideas, Beliefs, and Discourses: the Dispute Between Political Entrepreneurs and Defense Coalitions in Brazilian Agrotoxins Policy

Abstract

This research analyzed how advocacy coalitions competed for greater influence over Brazil's agrotoxins policy from 1999 to 2018. The premise was that changes in public policy are effected through changes in institutions, or the rules of the game. A theoretical approach was proposed that articulates the Advocacy Coalition Framework, New Discursive Institutionalism, and the Gradual and Transformative Institutional Change Model. A literature search and document analysis were conducted, emphasizing the usefulness of shorthand notes from meetings and public hearings in parliamentary committees, as well as interventions by parliamentarians in the National Congress. The collected data were then subjected to a content analysis based on a priori defined categories. Four hundred and sixteen individual and organizational agents were mapped, forming three advocacy coalitions defined by shared ideas: agroproductivists, agroecologists, and technocrats. The work of political and institutional entrepreneurs and policy brokers linked to the advocacy coalitions during this period was also characterized. Finally, the discursive strategies, behaviors, and types of gradual and transformative institutional change adopted by each advocacy coalition were described by analyzing key events in a timeline. The findings suggest that the entrepreneurs, as representatives of the coalitions, were interested in promoting changes to the scope of agrotoxins policy and the institutions that influence it. The main contribution of this research was outlining the coalitions' strategies and understanding the role of entrepreneurs as representatives of the coalitions.

Keywords
interest groups; new discursive institutionalism; gradual and transformative change; agrotoxins policy

Resumo

A presente pesquisa analisou como as coalizões de defesa disputaram por maior influência na política brasileira de agrotóxicos ao longo de 20 anos, entre 1999 e 2018. Foi estabelecido como premissa que as mudanças nas políticas públicas são efetivadas por meio de alterações nas instituições, entendidas como as regras do jogo, propondo-se uma abordagem teórica que articula o Advocacy Coalition Framework, o Novo Institucionalismo Discursivo e o Modelo de Mudança Institucional Gradual e Transformativa. A partir de busca na literatura e análise de documentos, com ênfase na utilidade de notas taquigráficas de reuniões e audiências públicas em comissões parlamentares e intervenções dos parlamentares no Congresso Nacional, os dados coletados foram submetidos à análise de conteúdo a partir de categorias definidas a priori. Mapeou-se 416 agentes individuais e organizacionais, constituindo três coalizões de defesa, definidas a partir de ideias compartilhadas: agroprodutivistas, agroecologistas e tecnocratas. Também foi caracterizada a atuação de empreendedores políticos, institucionais e policy brokers vinculados às coalizões de defesa nesse período. Por fim, analisando os eventos-chave em uma linha do tempo, foram descritas as estratégias discursivas, os comportamentos e os tipos de mudança institucional gradual e transformativa adotadas por cada coalizão de defesa. Os achados permitem inferir que houve interesse dos empreendedores, como agentes representativos das coalizões, em promover mudanças no escopo da política de agrotóxicos e das instituições que a condicionam. A principal contribuição desta pesquisa foi traçar as estratégias de atuação das coalizões e compreender a figura e o discurso do empreendedor como representativa das próprias coalizões.

Palavras-chave
grupos de interesse; novo institucionalismo discursivo; mudança gradual e transformativa; política de agrotóxico

Introduction

Recent studies have revealed a gap in the literature on the Advocacy Coalition Framework (ACF) concerning the possession and use of advocacy coalition resources (Ma & Vieira, 2020; Vieira, 2021; Ma, Lemos, & Vieira, 2020; Jenkins-Smith et al., 2018; Weible & Ingold, 2018; Weible et al., 2022). Vieira (2020), for example, points out that discourse can be a resource available to coalitions seeking changes in public policy. To this end, Vieira proposes a rapprochement between the ACF and New Discursive Institutionalism, the latter of which prioritizes the role of ideas and the discourse of agents. In other words, the argument is made that the concepts of ideas, beliefs, values, discourses, and institutions converge, building on the contributions of the Gradual and Transformative Institutional Change Model (Mahoney & Thelen, 2010; Streeck & Thelen, 2005). The aim is to address two gaps in the literature: first, the empirical applications of the ACF are examined in combination with other theories (Jenkins-Smith et al., 2018; Weible, 2018; Vicente, 2015; Ma et al., 2020); second, the model's application outside the original American pluralist democratic context is explored.

Keeping this in mind, the following research question was established: How do public policy entrepreneurs, who are agents of advocacy coalitions, use discourse to influence the course of a given policy according to their interests and values, thereby promoting changes in the institutional context?

To answer this question, an instrumental case study (Stake, 2011) was conducted on the National Agrotoxins Policy, which was established by Law 7,802 on July 11, 1989, and is also known as the Agrotoxins Law. This policy was modified by a number of subsequent decrees and complementary laws. The analysis focused specifically on the evolution of this policy over 20 years (1999 to 2018) and considered the behavior of entrepreneurial agents and defense coalitions in processes resulting in gradual and transformative institutional changes prior to the Bolsonaro government (2019 to 2022). This period was marked by large-scale changes and ruptures.

This study aimed to contextualize Brazil's agrotoxins policy during the period analyzed. It identified the active participants and entrepreneurs, the advocacy coalitions within the scope of this policy, and the strategies they adopted to achieve their goals. The study also examined discourse as a powerful instrument for spreading ideas and influencing institutions. Taking an integrative approach to different concepts and theories aims to generate insights that enable a better understanding of agrotoxins policy and the institutional environment that conditions it. In summary, the research results focus on identifying coalitions and public policy entrepreneurs in the subsystem based on shared ideas and beliefs, analyzing their behavior to bring about institutional change and policy reconfiguration, analyzing the ideas that constitute discourse, and characterizing discourse as a resource for entrepreneurs and coalitions.

The advocacy coalition framework and new discursive institutionalism

The ACF is a model that helps us understand the processes and outcomes in a given political subsystem. It is based on three central concepts: 1) advocacy coalitions, 2) policy learning, and 3) policy change (Haar & Pierce, 2021; Nam, Weible, & Park, 2022). A central assumption of the ACF is that the values, attitudes, and behaviors of individual actors in political processes are influenced by a belief system (Hornung & Bandelow, 2022). These actors organize themselves into advocacy coalitions (Sabatier & Weible, 2007; Sabatier & McQueen, 2009). In this scenario, a wide variety of actors can directly or indirectly influence events within the scope of the subsystem. These actors include members of all levels of government, interest group leaders, scientists, consultants, citizens, and the media (Weible & Sabatier, 2009).

Therefore, beliefs are fundamental to understanding how individuals organize into advocacy coalitions and act within a political system. These coalitions structure themselves into a hierarchical system of three levels (Sabatier & Weible, 2007; Sabatier & McQueen, 2009): 1) deep core beliefs, 2) policy core beliefs, and 3) secondary beliefs (Sabatier & Jenkins-Smith, 1993). Deep core beliefs are the most comprehensive and stable, and are less susceptible to change. They are predominantly normative. Policy core beliefs are considered moderate and broaden the substantive and geographical scope of the political subsystem. Although they are resistant to change, they can be adjusted in response to agents' perceptions of the environment or after acquiring new information and experiences. Their specificity within the subsystem guarantees the formation of defense coalitions and coordination of activities between members. Finally, secondary beliefs comprise the most restricted and empirically based beliefs and are the most likely to change over time (Sabatier & Weible, 2007; Sabatier & McQueen, 2009).

Political subsystems are the basic units of analysis. They are organized within a broader scope of the political environment, which includes a greater diversity of topics over wide geographical areas. These subsystems are defined by relatively stable parameters, external events, and opportunity structures that ensure long-term coalition action. They are also limited by the available resources, short-term restrictions, and other events within the subsystem itself (Weible et al., 2009). Each subsystem enables actors to form two to five coalitions that share normative and causal beliefs and engage in coordinated activities over time. Each coalition holds its own resources and devises strategies to achieve its political objectives (Sabatier & Weible, 2007).

Vieira and Gomes (2014) point out that the ACF does not explicitly consider institutional changes resulting from the behavior of coalitions. The study of coalition behavior can address this limitation by aligning other theories with the ACF, including the Gradual and Transformative Institutional Change Model and New Discursive Institutionalism.

New Discursive Institutionalism understands ideas and discourse as crucial elements in the analysis of institutional and human behavior. This theory presents an alternative approach to exploring the intersection of the three original versions (Hope & Raudla, 2012; Raitio, 2013). This approach is more dynamic and focuses on agents' subjective interests (Schmidt, 2008, 20, 2010).

Ideas can be understood from three levels of analysis. The first level is policy solutions, which are considered the most general and capable of changing quickly. The intermediate level consists of program ideas, which can be changed slowly and gradually to adapt to challenging contexts. The deepest level consists of philosophical ideas, which are deep-rooted premises or ideologies that influence the design of solutions and programs and are difficult to change (Schmidt, 2008, 2011).

Schmidt (2008) points out that different approaches and perspectives within the social sciences have turned to the analysis of ideas, including the ACF model. Thus, the concept of defining defense coalitions based on agents' beliefs and values would reflect the ideas defended by their members. The ACF points to the three levels of ideas mentioned above as a hierarchical system of beliefs that condition the formation of defense coalitions: secondary beliefs, policy core beliefs, and deep core beliefs.

Ideas are expressed through discourse. The big question is how some ideas dominate the political arena to the detriment of others. Rather than focusing on what was said, the aim is to understand the context in which the discourse was transmitted – where, when, how, and why it was said – and its impact on the agents – who said what to whom (Schmidt, 2008). Understanding the discourse enables us to not only understand how institutions function and behave over time, but also to envision ways of operating them to better serve the interests of those within the institutional context. Thus, discourse "[...] plays a decisive role in the development of institutions" (Magalhães, Martins, & Resende, 2017, p. 177).

Public policy entrepreneurs, institutional entrepreneurs, and policy brokers

Political entrepreneurs are individuals or groups who invest resources (time, energy, reputation, and money) to promote an idea and obtain future benefits (Kingdon, 2003). According to Mintrom and Norman (2009), political entrepreneurs stand out in the policy formulation and change process due to their high degree of social acumen, which involves understanding other agents and engaging in political debates. They also focus on defining problems and promoting appropriate solutions, value their network of contacts and teamwork, and recognize the importance of developing coalitions in favor of the causes they defend. Additionally, they lead by example and carry out actions that reduce the risks decision-makers perceive regarding the ideas they defend.

They can be inside or outside the government structure. They are skilled negotiators who are experts on a particular issue. They maintain valuable political connections and represent other agents who defend the proposed ideas. They influence policy formulation to create a more receptive environment for their proposals. Their main characteristic is persistence to guarantee the promotion of their ideas and garner support from other agents (Kingdon, 2003).

On the other hand, the concept of the institutional entrepreneur, coined by DiMaggio (1991), emerged as a term to describe individuals or organizations that establish institutional arrangements to better serve their objectives. These actors employ resources and assume responsibilities to create new institutions or modify existing ones (Maguire, Hardy, & Lawrence, 2004). Various actors can assume the role of institutional entrepreneur, including individuals, organizations, civil associations, social movements, and networks that unite different agents with a shared goal (Hardy & Maguire, 2008). Thus, it provides an agent-focused view of how institutions emerge and transform (Hardy & Maguire, 2008).

Battilana et al. (2009) point out that the work of institutional entrepreneurs consists of a three-stage process: (1) developing a vision for change, (2) mobilizing people to garner support for their ideas, and (3) motivating supporters to provide the necessary backing to implement the desired change. Another factor contributing to the success of these individuals is their ability to identify (or create) opportunities to implement their ideas and visions (Hardy & Maguire, 2008). Finally, the institutional context itself, marked by uncertainties, problems, tensions, and contradictions, provides opportunities for these actors to present themselves as agents of change and achieve their goals.

However, for a group of actors – either isolated individuals, groups, or organizations – called policy brokers, the dominant concern is acting to keep the level of political conflict within certain limits within the context of the political subsystem and/or the defense coalitions. They mediate between different interests to try to achieve reasonable and feasible solutions to problems, often in the form of government action programs. These solutions are guided by their beliefs and the values shared by certain coalitions (Sabatier, 1988).

In some respects, the concept of a political entrepreneur is similar to that of a policy broker because both refer to individuals who influence public policy decisions. However, according to Christopoulos and Ingold (2011), these agents differ in the roles they play and the impact their actions generate in a network context. While policy brokers tend to be more collaborative and take actions aimed at the smooth running of the system as a whole, political entrepreneurs are more interested in the returns they can obtain for themselves.

The gradual and transformative institutional change model (GTIC)

The GTIC emerged as a response to the dominant view in the institutional field regarding the behavioral patterns of institutions. It explains how the stability of institutional patterns conditions the behavior of agents to allow the reproduction and maintenance of the current order (Mahoney & Thelen, 2010; Streeck & Thelen, 2005).

Gradual, transformational change tends to manifest as gaps between rules and their interpretations or applications. In a context of formalized rules, there is room for ambiguity in their interpretation and implementation by actors. Consequently, loopholes emerge that allow actors to alter their interests, perceptions, and preferences, raising questions about institutional continuity (Mahoney & Thelen, 2010).

When analyzing institutional change processes, two factors must be considered: the characteristics of the political context, such as veto possibilities for agents, and the institution's level of discretion in interpreting and/or complying with rules. These factors are essential to determining the dominant profile of the change agent that may emerge in the institutional context in question and the strategies they will adopt to achieve their objectives (Mahoney & Thelen, 2010). Thus, the model reconciles agency and structural factors (Vieira & Gomes, 2014).

The GTIC indicates four possibilities for change: 1) displacement, 2) layering, 3) drift, and 4) conversion. Displacement is characterized by replacing existing rules with new ones. Change through layering occurs when new rules are introduced to existing institutions, progressively altering their structure and influence on the behavior of actors. Drift occurs when changes in the external context modify the impact of existing rules, which formally remain the same. Finally, conversion results from agents' new interpretations and applications of existing rules due to ambiguity (Mahoney & Thelen, 2010).

In addition to identifying possible types of change, the GTIC assesses the dominant behavior adopted by agents, offering a more nuanced view than a simple division between winners and losers. By examining agents' intentions to preserve existing rules and their compliance with institutional rules, four types of agents of gradual change are identified, as shown in Table 1.

Table 1
Types of change agents and their behaviors

Types of agents	Does the actor seek to preserve the existing rules?	Does the actor comply with institutional rules?
Insurrectionaries	No	No
Symbiotics	Yes	No
Subversives	No	Yes
Opportunists	Yes/No	Yes/No

In summary, Figure 1 illustrates how different theoretical models combine to explain how advocacy coalitions operate ideas and discourses through political and institutional entrepreneurs and policy brokers to bring about change in public policies. According to the ACF, different advocacy coalitions can collaborate and compete within a public policy subsystem based on a system of shared beliefs. These coalitions are mobilized at different levels of coordination, operating resources, and strategies.

Figure 1
Conceptual model

In turn, the GTIC and the precepts of New Discursive Institutionalism expand our understanding of how the institutional context influences the development of public policies and the ability of defense coalitions and their entrepreneurial agents to act. Discourse is fundamental to understanding the factors that influence these agents. Since discourse is a valuable tool for expressing and defending ideas, it becomes an important means of advocating for the adoption of proposed solutions, persuading other involved actors to defend their interests, and influencing the direction of the institutional environment to allow for the implementation of their policies – even if it means advocating for changes to the current rules.

Starting from the context that public policy is a political subsystem encompassing a network of actors with diverse interests and beliefs organized into advocacy coalitions, it is understood that some actors tend to assume a more prominent position when defending ideas and implementing solutions. Political and institutional entrepreneurs fall within this definition, while policy brokers take a negotiating stance and mediate between existing interests. The ideas and solutions that these actors defend are transmitted through their discourse, reflecting the belief systems of the coalitions to which they belong. The actions of entrepreneurs and coalitions are characterized as strategic, revealing the agents' dominant behavior to achieve their objectives and trigger processes of change in policies and institutions if it is in their interest.

Method

The research employs a case study method supported by documentary and bibliographic research. A wide range of documents were used, including laws, draft laws, norms, and decrees that constitute an institutional framework for agrotoxins policy, as well as instruments that produced changes throughout its evolution. These documents also include shorthand notes from debates in the National Congress and transcripts of speeches made by parliamentarians from the rostrum of the Chamber of Deputies and the Federal Senate, in which they expressed their opinions. These sources also include proposals for change and criticisms of agrotoxins policy, as well as technical notes and reports from the National Health Surveillance Agency (ANVISA), the Brazilian Institute for the Environment and Renewable Natural Resources (IBAMA), and the Ministry of Agriculture, Livestock, and Supply (MAPA) regarding agrotoxins consumption.

The research period was defined as twenty years (1999 to 2018), in line with the premise of the ACF and GTIC, which presuppose a minimum ten-year period for investigating the studied phenomenon (Mahoney & Thelen, 2010; Sabatier, 1988; Sabatier & Weible, 2007; Weible & Nohrstedt, 2012; Capelari, Araújo, & Calmon, 2015; Weible et al., 2019). This period was chosen due to a limitation imposed by the legislative houses. The Senate adopted the codification of audio recordings of public hearings in shorthand notes in 1997, while the House adopted this practice in 1998. This information was obtained through contact with legislative technicians from both houses via email and telephone.

To collect the shorthand notes, we first determined which parliamentary committees to analyze. Since Brazil's agrotoxins policy is characterized by the tripartite relationship between MAPA (agriculture), IBAMA (environment), and ANVISA (health) to carry out technical analyses and grant registrations for agrotoxins authorized for use in Brazil, we looked at the speeches of the permanent parliamentary committees of the Chamber of Deputies and the Federal Senate, which are responsible for agriculture, the environment, and health issues. The chosen committees were: 1) in the Chamber of Deputies, we chose the Committees on Agriculture, Livestock, Supply and Rural Development; the Environment and Sustainable Development; and Social Security and Families; 2) in the Senate, we chose the Committees on Agriculture and Agrarian Reform; Social Affairs; and the Environment, Consumer Protection, and Supervision and Control. We also considered two temporary committees with divergent proposals to change the current legislation: the Special Committee on Bill (PL) 6299/2002, which regulates plant protection agents and is portrayed by the media and sectors of society as the "Poison Bill," and the Special Committee on Bill (PL) 6670/2016, which establishes the National Policy for the Reduction of Agrotoxins (PNARA). Both committees ended with the conclusion of the 55th National Congress (2015 to 2019), leaving bills approved for deliberation and voting in the plenary of the Chamber of Deputies.

The next stage was to select the relevant shorthand notes for more in-depth analysis. The main focus was identifying hearings where the subject of "agrotoxin" and/or "crop defense" was the central agenda or a topic of debate because it was linked to other issues. A total of 198 shorthand notes were selected, 133 of which proved useful. In total, 350 hours and 46 minutes of hearings were analyzed. Regarding the speeches delivered in plenary, an exploratory search was first carried out on the websites of the Chamber of Deputies and the Federal Senate. The terms "agrotoxin" and "defense" were entered in the "subject" field. On the Chamber's website, 412 speeches were obtained by searching "agrotoxin," and 448 were obtained by searching "defense." After excluding repetitions and conducting a preliminary analysis, we determined that 290 speeches were relevant to the research. In the Senate, the results were 78 speeches for "agrotoxin" and 35 for "defense"; 66 of these speeches were relevant to the research. We selected plenary speeches based on the relevance of the speaker's information and contribution to agrotoxins policy, including presenting bills. We also considered the speaker's position on the issue, changes to current institutional rules, and the participation of relevant internal or external agents.

We used content analysis with a priori categorization for the analytical treatment of the data, as shown in Table 2. We also used NVIVO 11 software to help infer information about the main words in the speakers' speeches and analyze word clusters relating coalitions by word similarity.

Table 2
Categories of analysis defined a priori

Political entrepreneur	Individuals, groups of individuals, or organizations; Engagement in the defense and promotion of ideas that favor them; Mastery of a particular issue; They represent the interests of other agents; Attention focused on trying to solve problems through proposed solutions.	Kingdon (2003)
Institutional entrepreneur	Individuals, groups of individuals, or organizations; They seek to influence the context in order to modify or replace existing institutions with new institutional arrangements; They present a vision of the need for change; They mobilize people and motivate their supporters.	Hardy and Miguire (2008)
Policy broker	Individuals, groups of individuals, or organizations; They act to mediate conflicts between coalitions and/or within coalitions, seeking solutions that are reasonable and possible to implement.	Sabatier (1988)
Defense coalitions	Informal alliances between groups of individuals and organizations; Coordinated action within the political subsystem; Sharing a set of beliefs, values, and common interests related to politics, manifested through ideas.	Sabatier and Weible (2007) Schmidt (2008)
Strategies adopted	Displacement, layering, drift, and conversion	Mahoney and Thelen (2010)
Agent behavior and coalitions	Insurgent, symbiotic, subversive, and opportunistic
Classification of ideas	Levels of ideas (policy solutions, programs, and philosophies)	Schmidt (2008, 2010)

The collected data were organized in a spreadsheet to achieve two goals:

1. Structure the timeline of the policy's evolution and highlight the main facts and institutional changes that altered the original policy formulation, as outlined in the original version of Law 7,802/1989.

2. Identify the relevant actors within the scope of the analyzed policy. Based on this organization, it was possible to verify which actors could be identified as entrepreneurs, the coalitions to which they belonged, and the relevant parts of each participant's speech that expressed opinions, criticisms, defenses of current legislation, suggestions for changes to the institutional framework, and positions for and against the use of agrotoxins in Brazilian agriculture. These explanations also covered the procedures linked to the registration and release of agrotoxins and the difficulties faced.

The technique of discourse analysis was also used as an aid on excerpts of agents' speeches collected in shorthand notes from hearings and plenary speeches. The main aim was to understand and identify the ideas expressed by the actors. According to Schmidt (2008), the success of a speech depends on the scope and manner in which it is transmitted to its target audience.

As we categorized the excerpts from the analyzed speeches, we sought to identify the actors and organizations that exerted the greatest influence on the policy's evolution. This process also allowed us to understand how different actors expressed their ideas, proposals, and solutions for political and/or institutional change, enabling us to distinguish between existing coalitions and the aspects that permeated the power struggle between them.

Search results

Based on this information, the different coalitions and key players linked to the purposes of this study were identified. The purposes of this study are to understand the ideas that permeate agrotoxins policy and institutional changes, and to build a timeline of the evolution of agrotoxins policy. These points are explained in more detail in the following sections.

Identification of defense agents and coalitions

The first stage of the research involved identifying the individuals who participated in the analyzed committee meetings, including parliamentarians and civil society representatives. Of these parliamentarians, 107 were identified in the shorthand note analysis as having participated in hearings and deliberative meetings, as well as contributing to the debate on agrotoxins policy. Additionally, analyzing speeches in plenary sessions of the National Congress identified 95 congressmen, bringing the total number of parliamentarians who participated in debates on Brazil's agrotoxins policy to 202 over the 20-year period covered by the research. Of these parliamentarians, 138 were federal deputies, and 64 were senators. Five senators had previously served as deputies and were counted in both positions.

Of the members from outside the National Congress, 182 represent sectors of civil society. These individuals include academics and researchers from various research institutes and universities; representatives from scientific, business, and producer associations; trade councils; private companies; social movements; international organizations; non-governmental organizations (NGOs); and labor unions. Another 36 individuals are affiliated with various public bodies, including federal executive branch officials, public agency and foundation representatives, government council members, Federal Public Prosecutor's Office (MPF) members, Labor Public Prosecutor's Office (MPT) members, Federal Court of Auditors (TCU) members, and Federal Public Defender's Office (DPU) members, as well as state government department representatives. Thirteen agents were linked to more than one type of organization; four had links to both the public and private sectors.

A total of 109 public and private organizations were represented at the hearings. Four scientific associations; 19 business or producer associations; three class councils; eight government councils; five private companies; 15 civil society entities from a wide range of backgrounds; four social movements; six non-governmental organizations (NGOs); one international organization; 24 different federal public organizations (seven of which are ministries); five state bodies; four workers' unions; and 11 universities.

Based on the premises of the ACF identified by Sabatier and Jenkins-Smith (1993) and Sabatier and Weible (2007), we examined the Brazilian agrotoxins policy as a subsystem in which the 416 identified agents acted over the analyzed twenty-year period. To this end, we sought to interweave the precepts of the ACF's hierarchy of beliefs with Schmidt's (2008, 2010) New Discursive Institutionalism's approach of levels of ideas, recognizing that identifying the ideas expressed by the agents allows one to understand the structuring of the belief system that unites the defense coalitions within the subsystem.

Thus, three advocacy coalitions were identified in the analyzed policy: two opposing coalitions – the Agroproductivist and the Agroecologist – and the Technocrat Coalition, which balances between them and mainly includes organizations linked to the executive branch.

Therefore, considering the structure of beliefs and main ideas, the composition of each coalition could be defined, as shown in Table 4.

Table 4
Composition of defense coalitions

Agroproductivist Coalition
AGROPEC Consulting; National Plant Defense Association - ANDEF; EMBRAPA Soya; Uruguaiana Rice Growers Association (RS); Sectoral Chamber of Agricultural Inputs; Mato Grosso Rural Research and Extension Company - EMPAER; Brazilian Potato Association - ABBA; Thematic Chamber of Agricultural Inputs - CTIA; Federation of Agricultural Engineers of the State of Paraná; Brazilian Chemical Industry Association - Abiquim; National Technical Commission on Biosafety - CTNBio; National Forum of Agricultural Health Executives - FONESA; Brazilian Association of Generic Defensives - Aenda; Confederation of Agriculture and Livestock of Brazil - CNA; Brazilian Fruit Institute; Brazilian Cotton Growers Association - Abrapa; Coordination of Specialized Chambers of Agronomy - CCEAGRO; Institute of Applied Economic Research - IPEA; Brazilian Corn Growers Association - ABRAMILHO; Brazilian Agricultural Research Corporation - EMBRAPA; Organization of Brazilian Cooperatives; Brazilian Association of Soybean Seed Producers - ABRASS; EMBRAPA Vegetables; National Union of the Plant Defense Products Industry - SINDIVEG; Association of Brazilian Soybean Producers - APROSOJA Brazil; EMBRAPA Environment; National Union of Agricultural Aviation Companies - SINDAG.
Technocrat Coalition
National Health Surveillance Agency - ANVISA; Brazilian Institute of the Environment and Renewable Natural Resources - IBAMA; Ministry of Mines and Energy - MME; Administrative Council for Economic Defense - CADE; Ministry of Agriculture, Livestock, and Supply - MAPA; Ministry of Agrarian Development - MDA; Federal Council of Engineering and Agronomy - CONFEA; Ministry of Industry, Foreign Trade and Services - MDIC; Ministry of the Environment - MMA; Federal Council of Medicine; Ministry of Health - MS; Department of Health of the Federal District; Federal Police Department - DPF; Ministry of Foreign Affairs - MRE; Federal Court of Auditors.
Agroecologist Coalition
Alliance for Adequate and Healthy Food; National Council for Food and Nutritional Security - CONSEA; Federal Public Prosecutor's Office - MPF; National Articulation of Agroecology; Family Farming Cooperative; Small Farmers' Movement; Advisory and Project Services in Alternative Agriculture; Federal Public Defender's Office - DPU; Landless Rural Workers' Movement - MST; Brazilian Association of Biological Control Companies - ABCBIO; Ekip Naturama; Indigenous Movement; Brazilian Agroecology Association - ABA; National School of Public Health of the Oswaldo Cruz Foundation; Agroforestry Task Force; Brazilian Association of State Technical Assistance and Rural Extension Entities - Asbraer; Fazenda Malunga; Food and Agriculture Organization of the United Nations - FAO; Brazilian Association of Collective Health - ABRASCO; Federation of Agricultural Workers of the State of Pernambuco - FETAPE; Attorney General's Office - PGR; Organic Agriculture Association; Brazilian Forum for Sovereignty, Food and Nutritional Security; Cerrado Network; Association of Organic and Sustainable Coffees of Brazil - ACOB; National Forum to Combat the Impacts of Pesticides; Pouso Alto Agroecology Network; Honey Exporters Association - Abemel; São Paulo Forum to Combat the Impacts of Pesticides and Transgenics; Tobacco Industries Union; Permanent Campaign against Pesticides and for Life; Osvaldo Cruz Foundation - Fiocruz; Union of Agricultural Research Workers; Central Única dos Trabalhadores - CUT; Greenpeace; Brazilian Society for the Advancement of Science - SBPC; National Commission for Agroecology and Urban Production - CNAPO; Humans Rights Watch; SOS More Cerrado; National Supply Company - Conab; Bioregional Institute of the Cerrado; Land of Rights; Brazilian Beekeeping Confederation - CBA; Brazilian Institute for Consumer Protection - IDEC; National Union of Family Farming and Solidarity Economy Cooperatives - UNICAFES; National Confederation of Agricultural Workers - CONTAG; Chico Mendes Institute for Biodiversity Conservation - ICMBio; Via Campesina; National Health Council - CNS; National Cancer Institute - INCA; World Wide Fund for Nature - WWF Brazil.

The work of political entrepreneurs, institutional entrepreneurs, and policy brokers

Linked primarily to the Agroproductivist and Agroecologist coalitions, the entrepreneurial agents identified in Table 5 and discussed in more detail below have advocated for the ideas of these coalitions and promoted solutions to the problems they face and/or changes to the institutional rules that affect them. Additionally, they have sought to establish a presence in various committees throughout the period covered by the study. Notably, some of them have emerged as entrepreneurial agents in the Special Committees PL n° 6,299/2002 and n° 6,670/2016.

Table 5
Political entrepreneurs, institutional entrepreneurs, and policy brokers

	Agroproductivist Coalition	Agroecologist Coalition
Political entrepreneur	Brazilian Association of Generic Defensives (Aenda); National Plant Defense Association (ANDEF); Confederation of Agriculture and Livestock of Brazil (CNA); National Union of the Plant Defense Products Industry (SINDIVEG); Senators Blairo Maggi and Katia Abreu.	Brazilian Agroecology Association (ABA); Brazilian Association of Collective Health (ABRASCO); Campaign Against Agrotoxins.
Institutional entrepreneur	Confederation of Agriculture and Livestock of Brazil (CNA); Senators Ana Amélia Lemos, Blairo Maggi, and Katia Abreu; Deputies Adilton Sachetti, Covatti Filho, Luis Carlos Heinze, Luis Nishimori, Tereza Cristina, and Valdir Colatto.	Deputies Alessandro Molon, Bohn Gass, and Nilto Tatto.
Policy broker	Ministry of Agriculture, Livestock, and Supply (MAP); Deputy Tereza Cristina.	Deputy Alessandro Molon.

Thus, seven civil society organizations and two senators stood out as political entrepreneurs. Representatives of Aenda, ANDEF, the CNA, and SINDIVEG were political entrepreneurs associated with the Agroproductivist Coalition. These entities, which are linked to productive sectors, have been active in Congress for a long time. They have advocated for simplified procedures for evaluating agrotoxins and for more flexible legislation. They have also proposed changing the term "agrotoxin" to "pesticides" or "plant protection agents" and have argued for changes to the tripartite relationship. They claim that concentrating registration evaluations in MAPA or a technical commission linked to this ministry would provide greater agility and efficiency in analyses. They also played a leading role in defending the tax incentives granted to the production sector. During the Hevicoverpa armigera plague, they advocated for mechanisms that would enable emergency action to combat it. This resulted in Law No. 12,873/2013.

Among the parliamentarians, Senators Blairo Maggi and Katia Abreu stand out as political entrepreneurs because they are leaders of the Agroproductivist Coalition. Senator Blairo Maggi proposed the idea that formed the basis for the committee's debates and report on Bill 6,299/2002. His name has become synonymous with the interests of replacing the Agrotoxins Law in both coalitions. Senator Katia Abreu, on the other hand, negotiated and defended proposals to simplify product equivalence analysis procedures and mechanisms that would relax the requirements for using unregistered agrotoxins on crops in emergencies.

Three civil organizations – the ABA, ABRASCO, and the Campaign Against Agrotoxins – have played a leading role in the Agroecology Coalition. They mobilized public opinion to raise awareness about the risks of excessive agrotoxin use and the need for less intensive use of chemical compounds in agriculture, as well as to advocate for greater incentives for organic production. ABRASCO's role as a political entrepreneur is highlighted by its presentation of the Legislative Suggestion that led to Bill No. 6,670/16. The other members of the Agroecology Coalition recognize ABRASCO's work as a basis and source of reference for defending their positions. The ABA has also played a historic role in the National Congress, defending the use of the term "agrotoxin" to designate chemical compounds used in agriculture, as adopted in the 1988 Constitution. Additionally, the ABA and ABRASCO played an important role in drafting Bill 6,670/16.

In contrast, institutional entrepreneurs focus on ensuring that institutional arrangements reflect their interests. To this end, they may modify or replace existing institutions, which are defined here as the set of rules governing a given policy. They also mobilize people and motivate supporters to promote their ideas (Hardy & Maguire, 2008).

These characteristics were mainly observed in the actions of parliamentarians linked to the Agroproductivist and Agroecologist coalitions. As it is the body responsible for passing laws and changing legislation, this may explain the concentration of these entrepreneurs in Congress, to the detriment of sectors linked to civil society or the government.

These agents were observed to exhibit institutional entrepreneurial behavior through factors such as prominence in debates, active engagement in proposing changes to legislation, and mobilizing people's awareness of the relevance of these changes. Overall, 12 parliamentarians were identified as exhibiting institutional entrepreneurial behavior, nine of whom were part of the Agroproductivist Coalition. This demonstrates the strength of the coalition and that its interests prevailed in the main changes made through the decrees and complementary laws that were approved.

In the Senate, Ana Amélia Lemos, Blairo Maggi, and Katia Abreu stood out by motivating their allies and garnering support for modifying the Agrotoxins Law to ensure a more efficient evaluation process. Senator Katia Abreu presented herself as a great defender of sectors linked to the production of agrotoxins with expired patents. She mediated debates with the federal government, resulting in Decree No. 5,981/06. This decree simplified evaluation procedures for these products.

However, the Chamber of Deputies saw more consistent action by institutional entrepreneurs linked to the Agroproductivist Coalition. For example, deputies Luis Carlos Heinze and Valdir Colatto argued that Law 7,802/1989 and other related provisions did not sufficiently address the sector's interests, as evidenced by the following speech:

The productive sector can no longer withstand this pressure, this "regressionism," these rules, resolutions, and ordinances, and all this other paraphernalia that bypasses this House. We're doing something, but I don't know what. I'm sincerely disappointed and ashamed of our cowardice (Deputy Valdir Colatto, 2010, at a hearing in the Agriculture Committee of the Chamber of Deputies).

Additionally, establishing special committees enabled other parliamentarians to act as institutional entrepreneurs. Congressman Covatti Filho, on the side of the Agroproductivist Coalition, was responsible for submitting the project that resulted in the original installation of the PL 3,200/2015 Commission (later designated as PL 6,299/2002 after the addition of Senator Maggi's matter on changes to Law 7,802/1989). To justify the need for new legislation, he included long-standing demands from productive sectors in the bill. For example, he sought to change the term "agrotoxin" to "plant protection agent" and establish a commission to evaluate these products linked to MAPA. Mr. Nishimori, who wrote the final report approved by the committee, included mechanisms in the text of the new legislation to speed up the analysis procedures. These include registering equivalent products within sixty days, creating a temporary special registration, and allowing the use of agrotoxins on crops for which they were not originally approved but have proven effective. He also changed the terminology to "pesticides." As chairwoman of the PL 6,299/2002 Commission, Tereza Cristina demonstrated greater alignment with the productive sectors by advocating for the modernization of current agrotoxin legislation.

Conversely, Bill 6,670/2016 can be seen as the Agroecology Coalition's response to the advances in the opposing coalition's interests represented by Bill 6,299/2002. Congressmen Alessandro Molon, Bohn Gass, and Nilto Tatto stood out by defending the coalition's interests and proposals within the special committees. Molon and Tatto, respectively elected chairman and rapporteur of the committee, took a committed position to ensure that the PNARA was approved and that the final text included mechanisms that would encourage organic farming and symbolize civil society's demands for a production model with less reliance on chemical inputs. Tatto's final report included points that would change agrotoxin legislation by imposing time limits on registrations, restricting the term to products and agents based on chemical compounds, and establishing agrotoxin-free zones. Additionally, both Molon and Tatto participated in the commission that evaluated Bill 6,299/2002. They adopted a critical stance toward the proposed changes and advocated against the relaxation of agrotoxin legislation. They stood out by defending the need for a production model in Brazilian agriculture that is less dependent on chemical inputs, and they fought against the change in terminology adopted in the law to refer to these products. This is evident in the following speech:

The very name of this commission loses its meaning when we consider the harm that agrotoxins and agrochemicals cause to health. We must be careful with this issue, not least because it's guaranteed in the Constitution [...]. I'd like to draw your attention to this issue. In fact, the name change is intended to hide the fact that we're dealing with an agrochemical that's harmful to health (Deputy Nilto Tatto, 2016, at a hearing of the Special Committee on Bill 6,299/2002 of the Chamber of Deputies).

In addition to parliamentarians, the CNA's work included elements that classify it as an institutional entrepreneur. It did more than defend the need for changes in the law; it presented and advocated for proposals that benefited the productive sectors. For example, it supported the changes introduced by Decree No. 5,981/2006, which simplified the registration process for equivalent products, as well as the reduction in agrotoxin taxes in the country.

He also sought to disseminate the view among parliamentarians that these changes should be structured with the understanding that agrotoxins are an important product within the agricultural production chain. Therefore, the institutional arrangements of the policy should reflect a greater dimension of links with agricultural policy. He also pointed out the difficulties in establishing competencies between federal entities as a deficiency in the current legislation. This deficiency could only be resolved through a new law, as can be seen in the following excerpt:

More than 15 years have passed since the sluggishness was introduced, and nothing has changed. Innovating in business is unfeasible in this world. Another issue is the proper distribution of powers between federal entities. According to the Constitution, it's the responsibility of the Union, the States, and the Municipalities to legislate production and consumption. This creates a gray area: How far do the federal and member states go? It would be interesting to make this very clear in a new regulatory structure. This is fundamental to the structure's proper functioning. It's interesting that this is in a law and not a decree, ordinance, or normative instruction because it must be a permanent program (R. Minaré, CNA representative, 2017, at a hearing in the Agriculture Committee of the Chamber of Deputies)

In the context of agrotoxins policy, the polarized environment, in which positions are firmly held by dominant coalitions and are not open to changes in proposals, has not been conducive to the work of brokers. A general analysis of the collected data did not reveal actors who have consistently played this role over time. However, MAPA can be identified as an actor that occasionally mediated the different interests at stake. Although MAPA's agenda was more in sync with the interests of productive sectors linked to the Agroproductivist Coalition, it was able to recognize the demands of the Agroecologist Coalition. MAPA also stressed the importance of encouraging the use of biological compounds and stimulating the cultivation of organic products within agricultural production systems. This aspect is highlighted in the speech of its representative, Luís Eduardo Rangel:

Obviously, we've also been encouraging registration because organic farming is plagued by pests, a natural part of the agricultural model. Organic farming needs tools geared toward its reality. A lot of work is going into building models for organic farming inputs that are supported by the same legislation as agrotoxins [...]. There's a strong trend towards changing the profile of unregistered products. Many of the innovative products registered today, for example, are accepted in organic farming. Initially, they weren't proposed for organic farming, but they're accepted because their profile is so positive and low risk (Luís Eduardo Rangel, MAPA representative, 2014, at a hearing in the Agriculture Committee of the Chamber of Deputies).

Additionally, in the context of the special committee work, congressman Alessandro Molon and congresswoman Tereza Cristina, the respective chairs of the committees that assessed Bills 6,670/2016 and 6,299/2002, had to conduct their work in a manner that would mediate the different interests within the committees. This was necessary to ensure the work was completed and the final reports were approved. Other parliamentarians highlighted the importance of their work in ensuring the committees' progress. This can be seen in the following speech by the congresswoman:

There are two very distinct positions: some are for it and some are against it. We need common sense and balance to debate the issue. I want to be as democratic as possible. However, I ask for everyone's cooperation so that I can be flexible. Otherwise, I'll have to follow the Rules of Procedure. What I'd like is for us to debate the issue today with both sides participating at their leisure. This will allow us to move towards the best possible outcome for this very important issue. (Deputy Tereza Cristina, 2018, at a meeting of the PL 6,299/2002 Special Committee).

Brazil's agrotoxins policy – timeline

Brazil has solidified its standing as one of the world's leading agricultural producers. Brazil's climate, large tropical territory, and modernized production methods have enabled it to increase productivity per hectare, establishing itself as a leading food producer (Pignati et al., 2017; Silva et al., 2023). The advance of the so-called "Green Revolution" in the countryside – based on the intensive use of technology and inputs, as well as the monoculture of products such as corn, cotton, sugarcane, and above all, soybeans – combined with a tax policy favorable to large input producers and the granting of credit to farmers, has consolidated a model of agricultural production in which the use of agrotoxins plays a crucial role in Brazil's agricultural sector (Franco & Pelaez, 2016; Ollinaho, Pedlowski & Kröger, 2023).

In a 2015 dossier, researchers affiliated with the Brazilian Association of Collective Health (ABRASCO) noted that the average consumption of agrotoxins per hectare of planted area increased throughout the 2000s. The average increased from 10.5 liters per hectare (l/ha) in 2002 to 12 l/ha in 2011, a 14.28% increase. The following factors were identified as reasons for this increase:

[...] the expansion of GM soya planting, which increases the consumption of glyphosate, the growing resistance of weeds, fungi, and insects, requiring greater consumption of agrotoxins and/or the increase in crop diseases, such as Asian rust on soya [...]. An important stimulus to consumption comes from lower prices and the absurd tax exemption on agrotoxins, which makes farmers use more per hectare (Carneiro et al., 2015).

As Table 6 shows, until the 1980s, Brazil had no federal-level legislation regulating agrotoxins. The 1989 Agrotoxins Law was considered cutting-edge legislation on the subject. The formulation process reflected the diverging positions of two interest groups with different approaches. One group was more focused on the economy and agricultural expansion and more favorable to the use of chemical compounds. The other group was critical of the large-scale use of these products and concerned about their effects on human health and the environmental and social costs to the country (Brasil, 1989).

Table 6
The historical process that resulted in the Agrotoxins Law (Law No. 7,802/89)

Date	Events	Purpose
1982	Approval of the Agrotoxins Law in RS (State Law nº 7,747/1982).	To set guidelines for the distribution and marketing of agrotoxins in the state of Rio Grande do Sul and for holding those involved in the process accountable. It was the first of its kind at state level in Brazil and inspired the creation of similar legislation in other states in subsequent years.
1984	Preparation of a new draft law on agrotoxins by the Figueiredo government.	To replace the current legislation, dating from 1934. However, this proposal was not well accepted in Congress and was eventually withdrawn from the agenda by the Executive.
1985 / 1986	Preparation of a new draft law on agrotoxins, now under the Sarney government	To respond to the growing social demand for stricter legislation on agrotoxins policy.
1988	Launch of the Our Nature Program	Response to national and international pressure for greater government commitment to environmental protection, as a consequence of the murder of Chico Mendes.
April/1989	Urgent submission of the draft bill on agrotoxins to Congress	To take advantage of the "window of opportunity" by submitting the proposal for an agrotoxins policy to parliamentarians.
June/July 1989	Approval in Congress of Bill 1,924/1989	To approve the bill to create the Agrotoxins Law.

Although environmental groups were initially successful, it was not long before groups linked to the productive sector began to express their discontent. These groups took action to introduce changes to their scope, as seen in Table 7.

Table 7
Evolution of the regulatory framework governing Brazil's agrotoxins policy

Date	Events	Purpose
07/11/1989	Presidential sanction of Law No. 7,802/89, known as the Agrotoxins Law.	As the first comprehensive and specific law on agrotoxins, it introduced a new regulatory concept and guided the development of a new regulatory system for agrotoxins in Brazil.
01/11/1990	Presidential Decree No. 98,816/1990, to regulate Law No. 7,802/1989.	Consisting of 121 articles, the law regulated various points set out in Law 7,802/89. However, the agricultural sector and the agrotoxins industry criticized it.
1996 - 1998	Creation of GMC Resolutions 48/1996, 87/1996, 149/1996, 156/1996, and 71/1998, within the Mercosur framework.	It proposes harmonization of simplified procedures for registering technical products and products formulated by similarity. It establishes a list of active substances and formulations that are freely marketable among the member countries of the bloc. These substances and formulations should be registered in each member state with bodies pre-designated by Resolution 86/96. In the case of Brazil, the designated body is the Ministry of Agriculture.
06/06/2000	Sanctioning of Law No. 9,974, which amends the provisions of the Agrotoxins Law.	It established guidelines for the disposal and recycling of agrotoxin packaging, as well as the information that should appear on product labels. However, it did not alter the most critical aspects of Law 7,802/89.
01/08/2002	Publication of Decree No. 4,074/2002, which regulates the provisions of the Agrotoxins Law.	The new regulations create changes to the procedure for registering agrotoxins and harmonize national legislation with that of other Mercosur countries. The law also establishes the Technical Advisory Committee for Agrotoxins, which brings together the three bodies responsible for granting registrations and the Civil House.
12/17/2002	Publication of Law No. 10,603/02, which provides for the protection of undisclosed information.	The law establishes the information to be protected, the implications of this protection, time limits for protection, and how the responsible bodies should act to preserve intellectual property protection.
07/23/2004	Sanctioning of Law No. 10,925/2004	It exempts PIS and COFINS taxes on the import and sale of agrotoxins.
08/26/2004	Publication of Decree No. 5,195/2004	It regulates the provisions of Law No.10,925/2004.
03/24/2005	Sanctioning of Law No. 11,105/2003 (Biosafety Law)	It regulates the issue of transgenics by establishing safety standards and inspection mechanisms.
09/22/2005	Decree No. 5,549/05, which amends the provisions of Decree No. 4,074	It amends the articles dealing with components characterized as inert ingredients and additives.
12/22/2005	Publication of Decree No. 5,630/2005	It replaces Decree No. 5,195/04, expanding the range of agrotoxins exempt from PIS and COFINS taxes.
12/06/2006	Publication of Decree No. 5,981/2006	It rewrites and adds provisions to Decree No. 4,074/2002. It simplifies the process of evaluating agrotoxins without reducing rigor. It establishes three phases to evaluate products by equivalence. If a product is approved in at least one phase, it is granted registration. This represented a victory for Brazilian agricultural interest groups.
12/28/2006	Publication of Decree No. 6,006/2006	It establishes an exemption from the IPI tax on the production of agrotoxins.
07/23/2009	Publication of Decree No. 6,913/09	It establishes a specific approach for phytosanitary products approved for use in organic farming.
10/02/2010	Publication of Law No. 12,305, which establishes the National Solid Waste Policy.	It requires the implementation of reverse logistics systems for the disposal of agrotoxin packaging.
12/23/2011	Publication of Decree No. 7,660/2011	It replaces Decree No. 6,006/06 and maintains the IPI exemption on the production of agrotoxins in the country.
10/24/2013	Publication of Law No. 12,873/2013	It authorizes the granting of temporary emergency authorization for the import of agrotoxins, eliminating the need for prior registration. It also authorizes regulatory bodies to prioritize the analysis of products that combat pests causing an emergency situation.
10/28/2013	Publication of Decree No. 8,133/13, regulating Law No. 12,873/2013.	It allows regulatory bodies to prioritize the technical analysis of agrotoxins applicable to controlling and eradicating the disease caused by the pest that triggered the phytosanitary/zoosanitary emergency.
12/29/2016	Publication of Decree No. 8,950/2016	The decree replaces Decree No.7,660/11, maintaining the IPI exemption on the production of agrotoxins in the country.

First, they joined forces with Brazil's Mercosur partners to advocate for harmonizing Brazilian legislation with the agreements approved within the bloc. This would simplify registration procedures based on similarity. Then, they advocated for the introduction of new mechanisms to supplement the principles set forth in Law 7,802/1989, particularly through new decrees and supplementary legislation. These changes simplified the procedures for evaluating agrotoxins and guaranteed tax exemptions on the marketing and production of these products in the country. Finally, amid an economic crisis and political turmoil in Brazil, groups associated with the productive sector introduced sweeping changes to the governing institutions by repealing the current law and replacing it with new legal arrangements.

Thus, it is clear that groups linked to the environmental agenda and representative sectors of civil society have had limited capacity to stop the advances that productive sectors linked to agribusiness have made in shaping policy. They did little to contain the institutional changes unleashed, except for specific aspects – such as the approval of Decrees No. 6,913/2009, which established procedures for designating products for exclusive use in organic farming, and No. 7,794/2012, which created PNAPO and, later, Pronara, which was never implemented.

These changes to the legislation were made to speed up the registration process. These changes stemmed from instruments such as Decrees 4,074/2002 and 5,981/2006, which amended the Agrotoxins Law to simplify procedures for evaluating product registrations by equivalence, and Law 12,973/2013, which allowed for the import of agrotoxins exempt from registration during phytosanitary and zoosanitary emergencies in the country. Complementary legislation guaranteed the granting and maintenance of tax and fiscal exemptions for producing and marketing these products. Together, these instruments introduced gradual changes to public policy on agrotoxins.

In the 2010s, a turning point was reached in the Brazilian environmental agenda. Demands from sectors that proposed weakening current legislation advanced (Neves, 2016), and the issue of agrotoxins was not spared. From 2015 onward, amid a political and economic crisis in the country, debates about agrotoxins policy gained new momentum in the National Congress. Seizing the opportunity, members of parliament associated with agribusiness and representatives of the agrotoxins industry intensified their criticism of the legislation, advocating for deeper changes. These efforts resulted in the drafting of Bill No. 6,299/2002, which was approved by the plenary of the Chamber of Deputies in February 2022. Since then, it has been under consideration by internal committees of the Federal Senate in the form of Bill No. 1,459/2022. This bill would introduce significant changes to the legislation, beginning with the repeal of Law No. 7,802/1989. The bill meets the sector's historical demands by changing the term "agrotoxin" to "pesticide," authorizing the use of agrotoxins on crops for which they were not originally approved, and creating a temporary register while products are evaluated by competent bodies, as well as a unified registration system.

In response to this movement, sectors of civil society critical of agrotoxin consumption joined forces to combat the increase in agrotoxin use in the country. Consequently, a commission was established to debate and approve Bill 6,670/2016. If approved by Parliament, the bill would establish the National Program for the Reduction of Pesticide Use (PNARA) and create mechanisms to encourage the production of bio-inputs and agriculture with reduced agrotoxin use. The bill would also impose greater control on the use of these products, validate the registrations granted, and restrict the definition of agrotoxins to chemical inputs only.

Defense coalitions' strategies for institutional change

By analyzing changes in agrotoxins policy over the analyzed period from the perspective of the GTIC model, we can better understand the behavior of the coalitions and the entrepreneurs linked to them, as well as the impacts generated by the processes of institutional change. Table 8 summarizes the strategies and behaviors adopted by the coalitions, which are explained in more detail below.

Table 8
Institutional change strategies and coalition behavior

Defense coalitions	Type of strategy developed	Adopted behavior
Agroproductivist	Most of the time: layering Recently: displacement	Most of the time: subversive Recently: insurgent
Agroecologist	Most of the time: drift Recently: layering	Most of the time: mutualistic symbiotic Recently: subversive
Technocrat	Conversion	Opportunistic

In its process of establishing itself as the dominant group, the Agroproductivist Coalition mostly employed a layering strategy. This strategy involved defending the need to introduce additional mechanisms to the current policy, though not necessarily the immediate need to completely change the Agrotoxins Law, in order to make it more representative of their interests. Behaviorally, it adopted a subversive stance typical of agents who use layering strategies. Although it agreed to submit to and comply with the existing rules, it showed no commitment to them, advocating the need for new legal arrangements.

Later on, however, especially with the advent of the commission that evaluated Bill 6,299/2002, it changed its strategy to displacement. It became in its interest to replace the Agrotoxins Law with new legislation that it believed would be more modern and allow for greater effectiveness and efficiency in evaluating agrotoxin registrations. In the process, the Commission also changed its behavior, adopting an insurgent stance. In addition to losing interest in preserving the current rules, the Commission began acting to eliminate them by introducing a new institutional arrangement.

The Agroecologist Coalition took a different stance from the Agroproductivist Coalition because maintaining the legislation that conditions agrotoxins policy was in its interest. Although the coalition recognized some of the problems, it never spoke out in favor of replacing the legislation. Thus, to preserve the legislation, the coalition tightened its mechanisms in light of external events, such as the increased use of agrotoxins on Brazilian crops, reports of worker and rural community contamination from handling these products, and scientific research associating agrotoxins with an increased risk of human health disorders and environmental contamination.

This signals the adoption of a drift-type strategy since the sole objective was to alter the impact of existing rules to impose greater rigor on marketing and inspection. However, they behaved like mutualistic symbiotic agents without changing the rules, working for the stability of the system so that it would continue to benefit their interests.

However, the latest advances in the pro-change agenda of the Agrotoxins Law and the actions of sectors linked to the Agroproductivist Coalition may have motivated a change in the coalition's strategy. The coalition began to adopt layering strategies. The first step was collaborating with Dilma Rousseff's government (2011–2016), which led to the decree establishing the National Policy for Agroecology and Organic Production (PNAPO). This new strategy materialized with Abrasco's presentation of a people's bill proposal, which resulted in the commission that evaluated Bill 6,670/2016. While the coalition continued to support maintaining the Agrotoxins Law and its regulations, it realized the need to act more proactively to introduce new legal mechanisms. These new mechanisms would complement the current legislation and guarantee the preservation of its interests. During this time, the coalition's agents began to act as insurgents.

Finally, regarding the Technocrat Coalition, it took a stance of defending the current institutional rules that influence agrotoxins policy while recognizing the need to accept any changes approved by Congress. While some members, such as MAPA, have defended the need for changes to the Agrotoxins Law, the coalition itself has not actively supported these proposals. The coalition restricted itself to accepting the changes if they were implemented as a result of the displacement and layering strategies adopted by the other coalitions. Consequently, the coalition did not oppose the changes introduced over the last twenty years, which mainly served the interests of the Agroproductivist Coalition. Its strategy was to ensure the correct application of existing rules, even adopting a new interpretation, which was akin to a conversion-type strategy. Their behavior can be described as opportunistic because it was generally characterized by ambiguity regarding the preservation of existing institutions and compliance with the rules.

Conclusion

Based on an analysis of the period between 1999 and 2018, three advocacy coalitions were identified in the Brazilian agrotoxins policy subsystem: 1) technocrat, 2) agroproductivist, and 3) agroecologist. Their members were mapped, and the work of their political and institutional entrepreneurs and policy brokers was highlighted. The study observed how ideas and beliefs, when operationalized in discourse, lead to gradual and transformative institutional change – that is, coordinated efforts to alter, replace, reinterpret, or add to norms.

Returning to the research question, it is evident that public policy entrepreneurs leverage their position of relevance within the policy context and their legitimacy in representing a larger group of stakeholders to assert their interests more effectively. Through this, they can operationalize their ideas to garner more support and influence the direction of the policy, promoting the necessary changes in the institutional context.

However, it must be acknowledged as a limitation of the study that an agent's discourse in public environments can be ordered in the sense that it does not necessarily express their actual thoughts or beliefs on a given subject. Rather, it reflects the position of the organizations they represent or is polished out of fear of the audience's reaction. Additionally, the shorthand notes do not allow us to identify moments of hesitation or prolonged interruptions in the agents' speeches, which could indicate that they were unsure of what they were saying.

Regarding future research, we first suggest a more detailed examination of how the agents in the coalitions instrumentalize discourse. For example, one could conduct an analysis based on the modes of operation of ideology (Thompson, 1995). Additionally, the categories proposed by Schmidt (2008) can be used to characterize types of discourse. Furthermore, it would be worthwhile to investigate the relevance of new categories for characterizing gradual and transformative institutional change in the Brazilian context. After all, this is a democracy where laws are often disobeyed.

However, this research only used the ACF to map the coalitions. Therefore, it is worth investigating how these coalitions establish coordination within and between themselves, how learning may have caused changes in policy, how resources and strategies were used, and whether external events influenced the observed changes over time.

Approaching the same case through other theoretical lenses focused on agenda formation is also interesting. For example, the Multiple Flows Model considers the public policy entrepreneur as the protagonist, and Punctuated Equilibrium Theory offers an alternative explanation for institutional changes based on moments of rupture. In this sense, this approach might be useful for advancing the chronological analysis of the subsystem in question, considering the election of an extreme right-wing government in 2019. In other words, was this event a significant occurrence (Punctuated Equilibrium Theory) or an external shock (ACF)?

Finally, it is recommended that stakeholder analysis models be used to characterize the mapped actors more specifically, including the degree of influence of each stakeholder, resource sharing, and adoption of common strategies.

Table 3
Main ideas defended by the coalitions

Agroproductivist Coalition
Philosophical ideas / deep core beliefs	Agrotoxins are essential and indispensable to modern agriculture; The policy must provide mechanisms that speed up registration procedures and reduce barriers for productive sectors.
Program ideas / policy core beliefs	Changes in legislation should make the procedures for analyzing and registering agrotoxins in Brazil simpler and more flexible; The term "agrotoxin" has negative connotations and should be replaced in the legislation with "pesticide" or "plant protection agent"; Tax incentives granted to the sector protect production, considering the high costs of R&D for new products and the time it takes to obtain registrations; It is not possible to replace the use of agrotoxins with alternatives that do not involve chemical compounds on crops without compromising Brazil's position as a major global food supplier.
Policy solution ideas / secondary beliefs	Proper implementation of the policy is necessary to ensure greater security for the productive sector and higher crop yields; Factors related to Brazil's tropical climate justify the use of agrotoxins to combat pests and ensure productivity; Centralizing registration procedures in a single body would increase efficiency and food safety; The slow pace of regulatory bodies' actions is partly due to flaws in the existing institutional instruments; The agrotoxins used on crops are safe, resulting from significant industry investment in research and development; The smuggling of agrotoxins must be strongly combated as it is considered a major enemy of the production sector.
Agroecologist Coalition
Philosophical ideas / deep core beliefs	- The large-scale use of agrotoxins poses potential risks to human health and the environment; - The risks and harm caused by these products are the main reasons for a change in the Brazilian production model toward agriculture with fewer agrotoxins.
Program ideas / policy core beliefs	- Agroecology and organic farming offer an alternative to conventional agricultural practices; - We need an agricultural system that uses fewer agrotoxins and promotes sustainable development and bio-inputs; - Agrotoxins are negative and need to be discouraged in crops because they are characterized as "poison"; - We need to tighten agrotoxins policy to combat the large-scale, uncontrolled use of these products. This includes actions such as the aerial spraying of crops and the use of unauthorized chemical inputs; - Tax incentives granted to the agrotoxins industry must be eliminated.
Policy solution ideas / secondary beliefs	- Field workers are vulnerable to the risk of poisoning because they handle these products directly; - The population is susceptible to consuming food with agrotoxin residues, which can have harmful long-term health effects; - It is necessary to establish deadlines for agrotoxin registration renewals in the country; - The increase in agrotoxin consumption has not led to increased crop productivity. In fact, it is directly related to the use of transgenic seeds in the field.
Technocrat Coalition
Philosophical ideas / deep core beliefs	Laws and decrees should guide the actions of public bodies involved in public policy; Each body participating in the agrotoxin evaluation process must have the specific competencies necessary to ensure the safety of the analysis procedures.
Program ideas / policy core beliefs	Government actions regarding agrotoxins policy are based on the governing institutions, so the bodies responsible for policy regulation act in accordance with the legal principles established in current legislation; The current composition of the tripartite relationship between MAPA, ANVISA, and IBAMA is fundamental to evaluating agrotoxins, even though they occasionally disagree on some issues; It is necessary to improve infrastructure and human capacity to operationalize the registration and approval processes for agrotoxins.
Policy solution ideas / secondary beliefs	The use of agrotoxins is an issue that divides the coalition. Some members advocate for the use of these products in agriculture, while others support alternative agricultural production models that use fewer agrotoxins; The procedures adopted in the evaluations have a technical and scientific basis and are supported by the international community; The long registration evaluation queues are the result of insufficient financial resources and technical staff to meet the existing demand.

Acknowledgements

To the Coordination for the Improvement of Higher Education Personnel (CAPES).

References

• Battilana, J., Leca, B., & Boxenbaum, E. (2009). How Actors Change Institutions: Towards a Theory of Institutional Entrepreneurship. The Academy of Management Annals, 3 (1), 65-107. doi: 10.1080/19416520903053598.
  » https://doi.org/10.1080/19416520903053598
• Brazil. Law No. 7,802, of July 11, 1989 (1989). Provides for research, experimentation, production, packaging and labeling, transport, storage, marketing, commercial advertising, use, import, export, final destination of waste and packaging, registration, cl. Brasília: Presidency of the Republic, 1989.
• Capelari, M. G. M., Araújo, S. M. V. G., & Calmon, P. C. P. (2015). Advocacy Coalition Framework: a review of national research. Public Administration and Social Management, 7( 2), 91-99.
• Carneiro, F. F. et al. (2015). ABRASCO dossier: a warning about the health impacts of pesticides. Rio de Janeiro & São Paulo: Escola Politécnica de Saúde Joaquim Venâncio (EPSJV) & Expressão Popular.
• Christopoulos, D., & Ingold, K. (2011). Distinguishing between political brokerage & political entrepreneurship. Procedia - Social and Behavioral Sciences, 10, 36-42. doi: 10.1016/j.sbspro.2011.01.006.
  » https://doi.org/10.1016/j.sbspro.2011.01.006
• Dimaggio, P. J., & Powell, W. W. (1991). Introduction. In Powell, W. W.; Dimaggio, P. J. (Eds.). The new institutionalism in organizational analysis Chicago: The University of Chicago Press, 1-38.
• Franco, C. R., & Pelaez, V. (2016). The (Un)Construction of the Political Agenda for the Control of Pesticides in Brazil. Ambiente & sociedade, 19 (3), 215-232.
• Haar, R. N., & Pierce, J. J. (2021). Foreign policy change from an Advocacy Coalition Framework perspective. International Studies Review, 23 (4), 1771-1791. doi: 10.1093/isr/viab044.
  » https://doi.org/10.1093/isr/viab044
• Hardy, C., & Maguire, S. (2008). Institutional entrepreneurship. The Sage handbook of organizational institutionalism, 1, 198-217.
• Hope, M., & Raudla, R. (2012). Discursive Institutionalism and Policy Stasis in Simple and Compound Polities: The Cases of Estonian Fiscal Policy and United States Climate Change Policy. Policy Studies, 33 (5), 399-418.
• Hornung, J., & Bandelow, N. C. (2022). Party identification and cultural theory in Europe: methodologically advancing comparative studies of the Advocacy Coalition Framework. Journal of Comparative Policy Analysis: Research and Practice, 24 (2), 117-137. doi: 10.1080/13876988.2021.1891834.
  » https://doi.org/10.1080/13876988.2021.1891834
• Jenkins-Smith, H. et al. (2018). The Advocacy Coalition Framework: An Overview of the Research Program. In Weible, C. M., & Sabatier, P. A. (eds.). Theories of the Policy Process. 135-171, Boulder, CO: Westview Press.
• Kingdon, J. W. (2003). Agenda, Alternatives, and Public Policy (2nd ed.). New York: Harper Collins.
• Lieberman, R. C. (2002). Ideas, Institutions, and Political Order: explaining political change. The American Political Science Review, 96 (4), 697-712. doi: 10.1017/S0003055402000394
  » https://doi.org/10.1017/S0003055402000394
• Ma, J., Lemos, M. A. C., & Vieira, D. M. (2020). How is the Advocacy Coalition Framework Doing? Some Issues since the 2014 Agenda. Brazilian Journal of Political Science, 7-42.
• Ma, J., & Vieira, D. M. (2020). Learning and change in public policy: Exploring possibilities in the Advocacy Coalitions Model. Journal of Public Administration, 54, 1672-1690.
• Magalhães, I., Martins, A. R., & Resende, V. DE M. (2017). Critical discourse analysis: A qualitative research method Brasília, DF: Editora Universidade de Brasília.
• Maguire, S., Hardy, C., & Lawrence, T. B. (2004). Institutional Entrepreneurship in Emerging Fields: HIV/AIDS Treatment Advocacy in Canada. Academy of Management Journal, 47(5), 657-679. doi: 10.2307/20159610
  » https://doi.org/10.2307/20159610
• Mintrom, M., & Norman, P. (2009). Policy Entrepreneurship and Policy Change. The Policy Studies Journal, 37 (4), 649-668.
• Nam, A., Weible, C. M., & Park, K. (2022). Polarization and frames of advocacy coalitions in South Korea's nuclear energy policy. Review of Policy Research, 39 (4), 387-410. doi: 10.1111/ropr.12466
  » https://doi.org/10.1111/ropr.12466
• Neves, E. M. S. C. (2016). Institutions and Environmental Governance in Brazil: The Local Governments' Perspective. Journal of Contemporary Economics, 20 (3), 492-516.
• Ollinaho, O. I. Pedlowski, M. A. & Kröger, M. (2023). Toxic turn in Brazilian agriculture? The political economy of pesticide legalization in post-2016 Brazil, Third World Quarterly, 44 (3), 612-630. doi: 10.1080/01436597.2022.2153031
  » https://doi.org/10.1080/01436597.2022.2153031
• Pignati, W. A. et al. (2017). Spatial distribution of pesticide use in Brazil: A tool for Health Surveillance. Ciência & Saúde Coletiva, 22 (10), 3281-3293.
• Raitio, K. (2013). Discursive institutionalist approach to conflict management analysis: The case of old-growth forest conflicts on state-owned land in Finland. Forest Policy and Economics, 33, 97-103.
• 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. doi: 10.1007/BF00136406.
  » https://doi.org/10.1007/BF00136406
• Sabatier, P. A., & Jenkins-Smith, H. C. (1993). Policy Change and Learning: An Advocacy Coalition Approach Boulder, CO: Westview Press.
• Sabatier, P. A., & Weible, C. M. (2007). The Advocacy Coalition Framework: Innovations and clarifications. In Sabatier, P. A. (Ed.). Theories of the policy process (2nd ed.). Boulder: Westview Press, 189-220.
• Schmidt, V. A. (2008). Discursive Institutionalism: The Explanatory Power of Ideas and Discourse. Annual Review of Political Science, 11 (1), 303-326.
• Schmidt, V. A. (2010). Taking ideas and discourse seriously: explaining change through discursive institutionalism as the fourth 'new institutionalism.' European Political Science Review, 2 (01), 1-25. doi: 10.1017/S175577390999021X.
  » https://doi.org/10.1017/S175577390999021X
• Schmidt, V. A. (2011). Speaking of change: Why discourse is key to the dynamics of policy transformation. Critical Policy Studies, 5 (2), 106-126.
• Schmidt, V. A. (2008). Taking ideas and discourse seriously: explaining change through discursive institutionalism as the fourth 'new institutionalism'. European Political Science Review, 2 (1), 1-25.
• Silva, D. S. et al. (2023) Temperature effect on Brazilian soybean yields, and farmers' responses, International Journal of Agricultural Sustainability, 21 (1). doi: 10.1080/14735903.2023.2173370.
  » https://doi.org/10.1080/14735903.2023.2173370
• Stake, R. E. (2011). Qualitative research: studying how things work. Porto Alegre: Penso.
• Streeck, W., & Thelen, K. (2005). Introduction: Institutional Change in Advanced Political Economies. In W. Streeck & K. Thelen (Eds.), Beyond continuity: institutional change in advanced political economies (pp. 3-39). Oxford: Oxford University Press.
• Vicente, V. M. B. (2015). Public Policy Analysis from the Perspective of the Advocacy Coalitions Model. Revista de Políticas Públicas, 19 (1), 77-90.
• Vieira, D. M. (2021). Gradual and Transformative Institutional Change: The Action of Stakeholders and Defense Coalitions in the Case of the Belo Monte Hydroelectric Plant. Organizations & Society, 28, 398-421.
• Vieira, D. M., & Gomes, R. C. (2014). Gradual and Transformative Institutional Change: the Influence of Advocacy Coalitions and Interest Groups on Public Policy. Organizations & Society, 71, 679-694.
• Weible, C. M. (2018). Instrument constituencies and the advocacy coalition framework: an essay on the comparisons, opportunities, and intersections. Policy and Society, 37 (1), 59-73. doi: 10.1080/14494035.2018.1417705.
  » https://doi.org/10.1080/14494035.2018.1417705
• Weible, C. M., & Ingold, K. (2018). Why advocacy coalitions matter and practical insights about them. Policy and politics, 46 (2).
• Weible, C. M., & Sabatier, P. A. (2009). Coalitions, Science and Belief Change: Adversarial and Collaborative Subsystems. Policy Studies Journal, 37 (2), 195-212.

• Plagiarism check:
  O&S checks all documents approved for publication for plagiarism using a specific tool.
• Inclusive language:
  The authors use inclusive language that recognizes diversity, shows respect for all people, is sensitive to differences, and promotes equal opportunities.
• Data availability:
  O&S encourages data sharing. However, in compliance with ethical principles, it does not require the disclosure of any means of identifying research participants, fully preserving their privacy. The practice of open data seeks to ensure the transparency of research results, without revealing the identity of research participants.
• Funding:
  The authors would like to thank the CAPES for their financial support.

• Associate Editor:
  Maria Elisabete Pereira dos Santos

Data availability

O&S encourages data sharing. However, in compliance with ethical principles, it does not require the disclosure of any means of identifying research participants, fully preserving their privacy. The practice of open data seeks to ensure the transparency of research results, without revealing the identity of research participants.

Publication Dates

• Publication in this collection
  15 Sept 2025
• Date of issue
  Aug 2025

History

• Received
  19 Sept 2023
• Accepted
  07 Feb 2025