Special Collection: Mega-Projects, Contentious Action, and Policy Change in Latin America Policy Effects of Resistance against Mega-Projects in Latin America: An Introduction

In this introductory article, we present the special issue and outline our research agenda on extractive development, social mobilization, and policy impact in Latin America. We propose a shift in analytical focus from the study of resistance to studying the policy and institutional impacts of mobilization. We outline possible outcomes of interest and conditions contributing to the attainment of policy and institutional change. These conditions include movement characteristics – such as coalitions, repertoires, and alliances with state actors – and the socioeconomic, political, and ideational conditions that shape and constrain patterns of mobilization and the likelihood and durability of its impact. We also sketch the core themes and findings of the articles comprising the special issue, which cover sectors including mining, hydroelectricity, oil extraction, and accompanying infrastructural expansion across Central and South America. Several of the articles show how mobilization led to policy change while others caution against being overly optimistic about policy change without durable shifts in the structures that keep development models that prioritize the large-scale extraction of natural resources in place. We conclude by identifying pending questions and avenues for future research.


Introduction
The proliferation of protest and other forms of contentious collective action in response to mega-development projects has attracted growing attention, especially in relation to expanding extractivism in Latin America since the 1990s (Arsel, Hogenboom, & Pellegrini, 2016;Bebbington et al., 2008). 1 Much of the research on this contention focuses on local resistance movements, while another prominent line of inquiry examines the diffusion and effectiveness of mechanisms such as community consultation and compensation (D. H. Bebbington, 2012;Falleti & Riofrancos, 2017;Schilling-Vacaflor & Flemmer, 2015;Walter & Urkidi, 2017). The focus on the local leaves an important question relatively unaddressed. What impact, if any, does the proliferation of contentious action against mega-development projects such as mines and hydroelectric dams have on national policies and governmental institutional change?
The authors in this special issue analyse the consequences of social opposition to mega-development projects. By mega-development projects, we refer to large-scale forms of investment that characterize extractivist orientations toward economic development. Such projects often have local costs that can generate protest or broader patterns of contentious action. To understand how contentious action leads to political impact, we bridge analyses of community-based resistance to mega-development projects with the literature on the policy and institutional consequences of contentious action (Amenta, 2006;Bosi, Giugni, & Uba, 2016;Bosi & Uba, 2009;Giugni, 2004;Giugni, McAdam, & Tilly, 1999;Stearns & Almeida, 2004). Furthermore, we build on existing scholarship by placing social movement mobilization for policy and institutional change into a broader political economy framework. We explain why and how actors with various interests and power resources form contending coalitions around national policies and institutions relating to mega-development projects, paying attention to how economic, social, and political structures and ideational factors shape these contentious episodes. In the special issue, we present a series of articles that highlight the dynamics and mechanisms of mobilization as well as how the conditions in which mobilization occurs influence the policy and institutional impacts of social movements.
The issue of whether mobilization against mega-development projects has policy and institutional impacts has broad implications. A prominent feature of development policy in Latin America since the 1990s, and especially since the 2000s, has been the intensification of natural resource extraction. This trend holds for countries where the state plays a prominent role in economic development as well as for countries that continue to rely more heavily on private enterprise. Whether governments are of the left, centre-left, or conservative, extractive activities have been prioritized as a means to fuel economic growth. Increased state revenue also funds poverty alleviation programmes, especially following a consensus developed after 2000 that extreme inequality in the region is no longer tolerable (de Castro, Hogenboom, & Baud, 2016).
Yet since the 1990s Latin America has also seen enhancements, however uneven, in local community empowerment, citizenship rights, and indigenous and ethnic rights. These shifts have coincided with and influenced an upsurge in local resistance to intensified extractive activities by state enterprises and private corporations. Communities, often in remote areas, have resisted large-scale oil and natural gas projects, mining ventures, mega dams for hydroelectric energy, and agricultural expansion such as of genetically modified soybeans (Bebbington & Bury, 2013;Lapegna, 2016;Leguizamón, 2014;Sawyer, 2004), fearing these projects would destroy their ways of life due to negative environmental, economic, and social impacts, cultural disintegration, loss of local autonomy, and physical eviction. Given the nature of the conflicts, national and international environmental, human rights, and alter-globalization movement organizations have frequently joined anti-mega project campaigns, often as key coalition partners. For democratic polities, these developments raise questions about the responsiveness of government to citizen interests and preferences, and their commitment to the protection of rights. Social movements are perceived to be agents of change that raise issues generally excluded from policy agendas and public debate, or aim to activate policies that are only weakly enforced. To the extent that they engage with politics and the policy process, the policy consequences of movements are highly relevant. This is fertile ground for political science, political sociology, and political ecology.
In this special issue, the authors examine the varied types and degrees of movement impact on policies and institutional outcomes relating to mega-development projects, covering sectors including mining, hydroelectricity, oil extraction, and infrastructural expansion across Central and South America. Rose Spalding, Ludovico Feoli, and Eduardo Silva analyse how movements have influenced policy and institutional change in El Salvador, Costa Rica, and Chile, respectively; Paul Haslam investigates activism, institutional change, and policy feedback during the conflict around the Pascua Lama mine on the Argentine-Chilean border; Anthony Bebbington et al. and Denise Humphreys Bebbington et al. analyse regional trends in Central America and the Pan-Amazon region, respectively; and Lorenzo Pellegrini and Murat Arsel, in an "Explorations" piece, raise questions about the motives and consequences of social mobilization in the Northern Ecuadorian Amazon. A number of the authors examine how institutional politics combine with disruptive politics to push policy and institutional change forward, especially at the agenda-setting stage and occasionally in policy formulation. Meanwhile, others caution against being overly optimistic about policy change without durable shifts in the power structures that maintain extractivist orientations toward development in place. They remind us that progressive policy changes can be rolled back, weakening or even eliminating hardwon social and environmental safeguards. Sometimes mobilization may also not develop enough to exert any kind of pressure on policy and institutions, especially if the threat of repression is high, or if alternatives to extractive development have not been clearly articulated or lack legitimacy.
In the next section, we outline our analytical approach, first describing how we conceptualize social movements and movement outcomes in the context of extractive development and then turning to the question of how movements contribute to policy and institutional change in the context of structural and contingent factors that shape and constrain their collective action efforts. In the section that follows, we sketch the main contributions of the articles in the special issue and then conclude by reflecting on open questions and directions for future research.

Analysing impact: from opposition to movement outcomes
How do social movement actors with various interests and power resources form contending coalitions around policies and institutions relating to mega-development projects? How are their efforts shaped by underlying economic, social, and political structures and ideational factors? Building on existing understandings of contentious collective action (McAdam, Tarrow, & Tilly, 2001;Tarrow, 1994;Tilly, 1978), we focus on opposition to extractive projects that takes the form of public challenges to government action or claims for government action by social actors aiming to change the status quo with respect to mega-development projects. 2 When successful, such efforts secure increased protections for communities affected by extractive development, change the terms of debate by introducing new narratives and priorities onto policy agendas, introduce policies and institutions that regulate the impact of industry, or lead to more radical shifts in development models away from resource-extractive industries. In each instance, such changes affect the interests of opposing social groups and established authority. Social movements often seek to bring about long-term transformative change (Della Porta & Diani, 1999), yet they have both short-term and long-term goals, as well as unintended consequences.
Movement actors that mobilize around mega-development projects often involve coalitions comprising local residents and neighbourhood associations, environmentalists, indigenous organizations and communities, campesino organizations, women's associations, human rights activists, scientists, and academics along with their allies in political parties, state bureaucracies, and transnational activist networks (Bebbington & Bury, 2013;Carruthers & Rodríguez, 2009; articles in this issue). Movements use a range of repertoires to influence policy and institutional change, from protests and other forms of disruptive politics to collective action in institutional settings, such as lobbying elected representatives, electoral politics, and litigation. Activists operate at multiple scales, such as in local protests, national campaigns, and transnational advocacy networks. They direct demands to different kinds of actors such as subnational governments, national state agencies, and international organizations that influence domestic government actions. Mobilization may also move between scales, with consequences for movement impact. For example, sustained mobilization across localities may build to national-level campaigns that target national policies. Transnational advocacy networks may support national-level mobilization by lobbying international organizations, which in turn provide resources or put pressure on national governments (Keck & Sikkink, 1998;Moog Rodrigues, 2004;Silva, 2013). 3 By movement outcomes, we refer to the policy and institutional impacts of mobilization against mega-development projects. With respect to policy changes, we take an approach that considers impacts at different stages of the policy process and gradual effects over time (Amenta, 2006;Silva, 2015). Movements can have various impacts on politics: they can place issues on the policy agenda and shape policy formulation by providing inputs about the content of policy proposals. They can also create the conditions favourable to policy adoption by lobbying or attempting to shift public opinion to pressure representatives to adopt certain policies. Movements may have direct effects based on their specific policy demands; indirect effects whereby mobilization puts an issue on the political agenda but has no direct input into the policy formulation process; and mediated effects whereby movement actors work together with political allies and where the content of proposals may change in the process of negotiation (see Silva, 2018). 4 Past the stage of policy adoption, movements may also influence the course of policy implementation, such as through continued efforts to get favourable policies implemented and monitoring their implementation, pressuring for favourable policies to be enforced, or in the case of unfavourable policies, using delaying and diluting tactics. Possible policy outcomes of interest include the introduction of bans, moratoria, and oversight and regulatory mechanisms (including environmental impact reporting), as well as outcomes that expand environmental and indigenous citizenship rights such as participatory mechanisms based on the principle of free, prior, and informed consent and community compensation.
Social mobilization may also result in institutional changes, such as increased or decreased state capacity, as captured by changes in budgets, personnel and expertise, services offered, new bureaus, and the types of institutions regulating and overseeing extractive projects. Beyond regulatory agencies that handle environmental licensing and agencies tasked with questions of mining and national development, institutional settings where movement activity may have impacts include ombudsman offices, public prosecutors' offices, boards and commissions on issues relating to extractive industries, agencies responsible for consultation processes, and environmental tribunals. Some cases result in definitive policy shifts such as mining bans or nationalization, but we are also interested in the "layered" qualities of institutional change (Mahoney & Thelen, 2010) and incremental impacts. Rather than emphasizing gradual endogenous institutional change, however, we are especially attentive to the incremental ways in which "outsider" movement actors -sometimes in concert with "insider" institutional actors -can introduce changes to "the rules of the game" in ways that gradually reshape the institutional context in which mega-development projects operate. Institutional change can occur at various levels of policymaking (and government action can influence institutional change in the private arena also), though our primary focus here is on national-level policy and institutional outcomes.
How do movements achieve political change? For one, the findings in this special issue collectively suggest that coalitions connecting activists with state institutional arenas (Silva, 1994;Stearns & Almeida, 2004) matter for policy and institutional outcomes in struggles over mega-development projects. We treat the state as a disaggregated instead of a unitary actor, which implies that some parts of the state bureaucracy may be available for informal alliances with activists even as others promote policies favourable to extractive industries. Echoing findings from Giugni's (2004) joint-effects model and Amenta's (2006) political mediation model, several of the articles show that movements tend to be more effective when they have institutional access as well as political allies such as elected officials and state bureaucrats. For example, at the national level, connections to political parties can serve as bridging mechanisms that allow for movements to have direct influence in drafting the environmental policy agendas of candidates at election time, or by working directly with members of Congress to draft legislation (see Silva, 2018). Government agencies can also serve as sites where incremental changes in bureaucratic routines and practices in response to activist demands may be observed (Haslam, 2018). The case of the El Salvador mining ban shows just how different these political and bureaucratic forms of access are (Spalding, 2018).
Leadership, resources, and organizational structures -that is, a movement's "infrastructure" -also matter for the likelihood of impact (Andrews, 2001), as do ideational processes such as framing (Cress & Snow, 2000;Hall, 2010). For instance, movements that build ties and use compelling collective action frames across sectors and scales tend to have more reach (Spalding, 2018). Activist lawyers able to translate movement demands into legal claims in the context of legislative and judicial processes have been important, as have movement organizations' capacities to frame demands in ways that resonate with the broader public . Protest discourses developed at earlier historical junctures become important for later mobilization and outcomes, as in the construction of narratives about water as a national resource tied to the country's economic development (Feoli, 2018). Activists' abilities to identify and leverage opportunities provided by broader political developments can also contribute to increased access and impact. For example, when Chile was negotiating entry into the Organization for Economic Cooperation and Development (OECD), global environmental norms articulated through terms of accession to the OECD provided activists with more voice and impact, increasing their possibilities to shape environmental policies at the national level (Silva, 2018).
Looking across different stages of the policy process, it is important to pay attention to how the timing and sequencing of collective action matters for the extent to which movements achieve their short-term and long-term goals. For example, existing evidence based on the US case shows that movements are more likely to have direct impacts during the agenda-setting stage and are more likely to influence the passage of legislation when combining protest activities with institutional influence (Johnson, Agnone, & McCarthy, 2010). It is possible that movements protesting mega-development projects may similarly have a greater direct impact at earlier stages, and indirect impact in later stages if they lack access to the institutional spaces where policy is being made or pushback develops against their achievements. Movements' early impacts may also affect the trajectory and strategies of collective action; for example, small concessions in the short run may incentivize more mobilization in the hope of obtaining greater concessions, whereas blocked impact in the short run may result in movements looking for alternative institutional arenas and modifying their repertoires. Movements' early impacts may also consolidate their supporters and build broader coalitions.
Movements do not always have impact, however, nor are their impacts permanent. Waves of protest and political pressure may result in recognition and some responsive policies, which produce change, encouraging more mobilization and further concessions leading eventually to implementation. However, these tendencies may also be cut short, reversed, or met with repression and the criminalization of protest. Movements can achieve their goals, but they can also be co-opted, pre-empted, or ignored (Gamson, 1975). Policy impacts may therefore be episodic and not necessarily durable; cycles of protest and pressure by civil society actors may lead to one round of policy impacts, followed by implementation and a reaction by countervailing coalitions in favour of mega-development projects, leading to environmental and social safeguards being suddenly or incrementally rolled back (Humphreys . In most cases, the path from mobilization to policy and institutional outcomes is not a linear one. In the analytical framework developed in this special issue, power is relational. The processes we describe are not just about the power of movements and the coalitions they form; the ability of movements to contribute to policy and institutional outcomes is also about the power of opponents. As the power of movements waxes, the power of opponents may wane and vice versa. From this perspective, then, policy change may be the result of a shifting balance of power -or of the political settlement (Bebbington, Abdulai, Bebbington, Hinfelaar, & Sanborn, 2018) -between coalitions that support competing policies. Socio-economic, state, and ideational factors condition the dynamics of coalition formation and the power resources of contending coalitions in the policy process.
To begin with, socio-economic, state-political, and ideational factors, as well as their international dimensions, define many of the actors and their interests, policy stances, and power resources. For example, mega-projects tend to be supported by large-scale domestic and international business elites, and by state actors (including elected politicians) who rely on extractive enterprises for economic growth and fiscal resources to sustain state activities. The sources of their power and influence are economic as well as political, by virtue of controlling the locus of legitimate authority, and, ultimately coercion in society. Ideational factors can be a further source of support for the interests of these elites when, for instance, middle class groups and subordinate social groups value orthodox economic development more than potential social and environmental consequences for their societies or for specific populations such as indigenous peoples.
Opposition to mega-projects is most likely among those groups who are excluded from the political settlement and also experience material and ideational dissatisfaction with respect to how power is being wielded. While such opponents are often framed as subordinate social groups such as farmers, peasants, indigenous peoples, and those who live in communities directly affected by a megaproject, they may also be elites, professionals, and technocrats who are excluded from governing power. The power of such opposition depends on organizational and/or associational capacity and the ability to disrupt economic activity and political order, as well as the broader purchase of ideational factors related to valuing the environment and cultural integrity as elements of community and national wellbeing. Middle class groups may also be swayed by such ideas, and members of such groups are often the organizers of environmental nongovernmental organizations (NGOs), many of which may work on policy issues. Business groups active in environmentally-friendly industries, such as eco-tourism, alternative energy generation, or organic agriculture, among others, may also frown upon -or oppose -traditional megaprojects. Last, but not least, are state actors and politicians (including at the subnational level) whose ideals and/or whose organizational missions lead them to support environmentally friendly, grassroots-oriented and socially inclusive initiatives.
More or less stable competing coalitions involving these actors underpin the balance of power between social forces that support or oppose megaprojectdriven extractive development. The dominant coalition tends to include largescale national and international business, state and political authorities that control the most important sources of political power, many middle-class groups, and some subordinate social groups. The opposing coalition tends to include excluded social groups mentioned above, alongside national and international environmental NGOs, politicians and political parties whose platform supports progressive environmental policy, some heads and staff of usually second-tier state institutions, and socio-environmentally conscious middle-class groups.
Typically, the dominant understanding, or settlement, is that large-scale private business -or in some instances, as in Feoli's article in this issue, large-scale state enterprise -is responsible for producing economic growth and employment which are, in turn, viewed as the fundamental goals of development. In Latin America, extractive enterprises, and large-scale energy and infrastructure investments are the basis for such models of development. Thus, their economic power carries great political weight. Middle classes will support this style of development in the interest of improved standards of living as might subordinate social groups, especially if some of the tax revenue that is generated is ploughed into social programmes. Arsel, Hogenboom, and Pellegrini (2016) have called this the extractive imperative.
While such understandings are dominant, they are not necessarily hegemonic. Concerns for the socio-environmental consequences of mega-development are also part of policy debates, there is some international and domestic support for addressing these issues, and there is pushback in the form of contentious action. The question is, from a position of relative weakness, when and how can contentious action contribute to progressive policy change around these concerns and avoid rollback? The answer in part depends on shifts in the power balance -or settlement -between supporters of development-as-usual and advocates of environmentally and community sensitive reforms. Such shifts depend on changes in specific mixes of economic, political, social, and ideational power. They also depend on the abundance of the natural resource in question, its significance for the economy, and the availability of alternatives.
Power, however, is also relational. For coalitions opposing the expansion of mega-projects, building organizational, political, economic, and ideational power works best, all other things being equal, when supporters of the traditional developmental paradigm weaken. For example, international conditions may be more favourable, as when commodity prices fall or international organizations promote social equity and more environmentally conscious governance arrangements. The litigation of high-level corruption may weaken elites and arguments associated with mega-development, as in the case of the Lava Jato crisis that has affected Brazil and other countries in Latin America (Humphreys . New political forces that support progressive policies may take office. The significance of a commodity for overall economic growth is also relevant. If business seeks to rapidly expand exploitation of a commodity that is not significant for economic growth and whose extraction has not yet been developed, governments faced with strong resistance and its political consequences may choose not to support such development (see Spalding, 2018, and. This is the political-economic context that shapes the contentious action of social movement organizations and their effect on policy. The connections they make with political elites and state bureaucracies, environmentally friendly business sectors, subordinate social groups and their organizations, and international organizations can augment their power. These connections shape the policy networks within which they move, the weight of public opinion that can be brought to bear on policymakers, and the degree and kind of disruption they can produce. Network connections inform whether contentious action might affect policy change by direct, mediated, or joint effect mechanisms.
The articles in this special issue explicitly or implicitly situate their subjects within this analytical framework. That said, progressive policy change, as we have pointed out, is not permanent. Rollback is always an option (as shown by Humphreys . As rollback occurs, social movement organizations lose power while shifts in economic, political, social, and ideational factors favour their opponents. It is, to some extent, a reverse process from the one described above. This special issue includes examples of this process as well.

Mapping the special issue: social movement impacts and policy change
The articles in this issue examine the policy consequences of collective action in response to mega-development projects, covering sectors including mining, hydroelectricity, oil extraction, and infrastructural expansion. The intensification of natural resource extraction has been met with social resistance, which in turn has pushed national policies and institutions to be more responsive to critiques of extractivist development, with a wide range of outcomes. For example, social movement groups contributed to the introduction of the mining ban in El Salvador, generated public debate about the assumptions of the hydroelectric model in Costa Rica, introduced environmental concerns onto the public agenda and contributed to strengthening environmental institutions in Chile, and generated bureaucratic change and legislation for the protection of glaciers in Argentina. In such contexts, coalitions of groups opposing mega-projects have affected not only the course of individual projects but also national-level policies and institutions. Yet in other cases, mobilization has had more limited impact. In the Pan-Amazon region, policy changes toward social and environmental protections (themselves secured by previous cycles of coalition building and pressure) have not been durable, and industrial-scale mining and hydrocarbon extraction is once again expanding, accompanied by major infrastructural investments. Policy change also does not necessarily entail revising deeply embedded extractivist orientations toward development, as in the case of a new law in Ecuador specifying labour inclusion for local residents without questioning dominant extractivist logics. In short, mobilization has influenced national policy and institutional change throughout Latin America, but this influence varies widely and the dynamics of the underlying balance of power in each country both shape the possibilities of collective action and threaten possible reversals to existing gains.
Studying one of the most notable recent policy shifts in the region, Rose Spalding analyses the process whereby El Salvador became the first country to adopt a national ban on all forms of metal mining. She traces how, over twelve years of mobilization, a growing mining resistance movement intersected with formal institutional spaces to bring about a major shift from policies favourable for mining investment to policies that stopped it. The case study shows that the movement played an especially critical role in articulating grievances related to mining and introducing them into national policy discussions, thereby influencing agenda setting. The Mesa Nacional Frente a la Minería Metálica, for example, brought together grassroots community organizations from the mining zone and national NGO allies with knowledge of environmental rights and legal protections, as well as building linkages with allies in the Catholic Church. Organizations affiliated with the Mesa sustained public interest in the mining question by organizing around the defence of water, and, once various legislative options to address mining were on the table, providing evidence of local opposition by conducting municipal-level consultations on mining in the country's gold belt. The analysis also shows how the targets of movement actors adapted their strategies to deflect attempts to influence policy, delay efforts to implement favourable policies, and block the demands of challengers. Spalding finds that activists were more likely to have input into how policies were formulated, adopted, and implemented when they developed alliances with institutional actors and identified spaces where public demands could be introduced into the formal policy process.
Using a historic lens to analyse debates about Costa Rica's energy sector, Ludovico Feoli shows how social movement groups are challenging the country's state-led electricity model premised on hydroelectric power. The most recent episode of mobilization has centred on the development of Diquís, a new 655-megawatt hydroelectric plant planned as the largest hydroelectric project in Central America and which the state utility company, the Instituto Costarricense de Electricidad (ICE), considers to be its key energy project for the next twentyfive years. Feoli argues that the policy and institutional impacts of the recent mobilization must be seen in terms of a longer history of collective action around the control of valuable energy resources that has contributed to defining the characteristics of the energy sector. Starting with Costa Rica's first sustained social movement in the electricity sector, which resulted in the nationalization of all electricity derived from hydraulic forces, the analysis shows how twentieth-century movements made demands relating to national energy sovereignty and the affordability of electricity that led to the institutional dominance of ICE. While past social movements defended ICE against privatization, emphasizing its role as a steward of the country's energy sovereignty, recent social movement actions directed at stopping public and private sector hydroelectric projects have shifted focus, keeping the emphasis on energy sovereignty but also critiquing the socioenvironmental impacts of hydroelectric mega-projects. Movement actors such as environmentalists, indigenous organizations, and local community organizations have emphasized greater openness, accountability, and environmental responsibility of the institutions involved in hydroelectric projects, as well as questioning the large-scale extractive energy model and opening a debate about the future direction of the national energy sector.
Anthony Bebbington and his co-authors take a regional perspective, investigating how patterns of large-scale natural resources and infrastructure investments affect forests and forest-dependent populations in Central America. Using an analytical framework grounded in the political economy of natural resources, political settlements, and the sociology and politics of policy change, they discuss different social responses to these investments, and their impacts on natural resource governance policies. The authors note that mining and dam-building elicit more collective action relative to energy or road access projects, while there is far less mobilization around forest loss. While the effects of protest are often constrained by political settlements committed to resource extraction, and the intimidation and repression of activists, social mobilization has sometimes led to policy change. Examples from El Salvador, Guatemala, Honduras and Panama suggest that favourable outcomes are facilitated by movement repertoires that include direct action, case-based and strategic litigation, research aimed at shifting public debate, and negotiations with investors and policymakers. In some cases, protest has taken a pragmatic form, as in efforts to secure institutional responses that allow for direct, community-based, and market-oriented management of forests, while other protest has been more contentious. Engaging the public and political elites through protest has provided legitimacy to national-level mobilization. Transnational ties have also been influential, and some of the cases played out with the involvement of international organizations such as the Inter-American Court of Human Rights. Even when protest has driven policy change, the authors note that the dynamics of the political settlement and the broader political economy constantly threaten reversals to such gains.
Turning to South America, Eduardo Silva analyses the consequences of the Chilean environmental movement for gradual institutional and policy change during the first twenty years of the country's transition to democracy. Tracing movement influence across four presidential administrations affiliated with the centre-left party coalition, the Concertación de Partidos por la Democracia (CPD), he analyses how social movement engagement with political parties and the policy process combined with protests to affect debates and outcomes relating to the environment. Party-movement connections provided environmental organizations with openings to influence the policy agenda of CPD governments by creating channels of access to actors in the executive and legislative branches. However, the dominance of business elites and the political bargain struck to make democratization possible led the first three CPD governments to support only weak environmental regulation, limiting the direct influence of movements past the agenda-setting stage. Silva embeds his analysis within a framework that makes explicit how national factors such as the balance of power among political and economic elites, as well as international factors such as a country's position in the global economy, shape the possibilities of contentious action to have direct, indirect, or mediated effects on policy. Silva examines major institutional changes such as the creation of the Ministry of the Environment as well as policy effects in the energy sector, looking at how environmental movements were able to put non-traditional renewable resources on the policy agenda as a result of protest politics and working directly with key parliamentarians from the legislature's lower chamber, the House of Deputies, to set targets regarding how much energy is to be produced by non-traditional renewable sources by 2020.
In "The Two Sides of Pascua Lama," Paul Haslam examines the consequences of mobilization against the emblematic Pascua Lama gold mine on the border of Argentina and Chile. He argues that the ability of activists to achieve outcomes that they value is conditioned by how bureaucracies implement policies and regulations, and compares mobilization and institutional responses on both sides of the border across three phases spanning from 2001 to 2018. His account shows how initial impacts on policy and institutional changes resulted in political and institutional feedback loops: state actors strategically responded to the changes, reshaping the configuration of political opportunities and the policy influence of movements over time. Nevertheless, where activists dealt directly with regulatory agencies, they were able to have influence. For example, in Chile, social mobilization at an earlier stage led to the creation of a government unit to conduct an inventory of glaciers and come up with a national glacier strategy, which led to a case-by-case review of different mining projects' impacts on glaciers. Later, activists actively pressured administrative institutions tasked with environmental protection to enforce compliance with social and environmental regulations. In Argentina, though a glacier inventory was conducted as part of new legislation to protect glaciers, establishing no-go zones for mining, the implementation of the law was limited and activists gained little traction. Haslam concludes that relatively minor changes to bureaucratic policies, regulatory enforcement, and judicial oversight can shape the influence of social movements over extractive sector governance.
Denise Humphreys Bebbington, Ricardo Verdum, and Cesar Gamboa analyse the governance of extractive and infrastructural development projects in the Pan-Amazon region. Moving beyond a focus on particular extractive sites to consider the spaces "in between", they note that natural resource extraction requires infrastructural investments in the form of roads, pipelines, railways, and port facilities. These different investments interact and reinforce each other to produce large-scale environmental and social impacts. They argue that development policy in the region increasingly revolves around promoting synergies among these different investments, and discuss recent government and private sector efforts to dilute or eliminate social and environmental safeguards protecting indigenous and conservation lands and to remove bureaucratic "impediments" to investment -a process they refer to as "roll back". From the mid-1990s to 2010, organized Amazonian groups supported by Brazilian NGOs, some government agencies, and transnational activist networks participated in episodes of social mobilization that led to institutional and policy changes that increased protections. However, many of the underlying rules of the game and ideas remained unchanged, and key parts of the infrastructural and extractive industry sectors continued to view indigenous people and protected areas as obstacles to growth. In recent years, earlier commitments are being destabilized by new laws and norms that aim to scale back environmental licensing procedures, restrict consultation processes, reduce the constitution of new indigenous lands, allow extractive operations in protected areas, and permit fast-tracking projects to create more favourable conditions for infrastructure and extractive development. While protest may drive institutional change, the durability of such changes depends on shifts in the balance of power and underlying political settlement.
Finally, in the "Explorations" section, Lorenzo Pellegrini and Murat Arsel analyse oil extraction and development in the Ecuadorian Amazon, focusing on contentious collective action in response to the socio-environmental impacts of extractive operations on the indigenous and mestizo populations of Orellana and Sucumbíos provinces. Pushing back against literature where cases of mobilization around environmental concerns are overrepresented, they argue that local reliance on existing mega-projects can constrain the demands made by social movements and generate contradictions, leading to claims for environmental remediation and respect for ancestral territories simultaneously with demands that implicitly require the extractive industry to expand and intensify operations. The article raises questions about how existing mega-projects create circular logics, whereby community groups and movement organizations demand a different distribution of rents generated through oil extraction, rather than challenging the oil industry altogether. Organized collective action rooted in environmental justice concerns thus coexists with demands that reflect and reinforce the dominance of extractivist approaches due to the unviability of alternative strategies of development.
Collectively, the special issue offers several lessons about the impact of mobilization on policy and institutional change. First, it shows that movement demands for change vary greatly, from mobilization around a mining ban in El Salvador to the introduction of inclusive labour policies for oil extraction projects in Ecuador. Next, it demonstrates the range of strategies and tactics that movement organizations and their networks use to pressure governments for policy and institutional change. Movement groups have worked through political parties and their connections to elected representatives and by pressuring and doing "activist monitoring" of bureaucratic agencies. They have used transnational advocacy networks to generate support and strengthen their impact. Some of the studies clearly show that the most powerful organized collective action has been that which effectively bridges multiple kinds of groups and makes demands not only through disruptive politics but also by working within formal institutional spaces.
Yet the analyses here also underscore that social movements face structural and ideational conditions that make policy and institutional change favouring social and environmental protections difficult to achieve. Powerful elite coali-tions backing mega-development projects across countries tend to view social and environmental safeguards as impediments to investment. Communities and activists, when faced with lack of access or the threat of repression, may refrain from mobilization, choose pragmatic coping strategies, or pursue other forms of organized collective action and claims making rather than pressuring for radical policy change. In some cases, they may also be divided as to what constitutes favourable policies, since extractive industries continue to be sources of employment even when communities recognize their destructive environmental and social impacts. Mobilization is therefore potentially powerful, but not uniform in its ability to generate national policy and institutional change.

Future avenues for research
In this introductory article, we have proposed a research agenda to identify when and how social movements mobilizing in reaction to mega-development projects engage with and influence political and institutional outcomes. Our analytical approach brings together theoretical and empirical insights from the expansive literatures on community resistance to mega-development projects and on social movement outcomes with a comparative political economy framework, allowing us to take into account movement characteristics and repertoires of collective action as well as a broader set of structural and contingent factors that shape the possibilities of contentious action and impact.
The articles in the special issue cover substantial ground. Yet there is further case study and comparative work to be done to analyse the consequences of social mobilization for policy and institutional impacts across countries as well as across types of extractive mega-development projects such as mines, dams, and capital-intensive agribusiness. A number of additional questions and possibilities for future research remain. First, while most of the articles here address legislative and bureaucratic changes, the expanding judicialization of policymaking and of individual megaprojects in some countries makes court rulings and their effect on policy and institutional change another interesting arena for research (for example, see Couso, Huneeus, & Sieder, 2010;Sieder, Schjolden, & Angell, 2005).
There is also more work to be done on outcomes at different scales. Here we have focused on national-level outcomes, yet policy and institutional change at the subnational and international scales also shapes models of development. How do policy changes at the subnational level influence national policies, and vice versa? Are international organizations responsive to resistance movements? Do they adjust norms, principles, procedures and policies? In addition to statebased outcomes, Haslam's and Feoli's articles in this issue also show the need to understand more about movement impacts on corporate practices and "private regulation" of extractive industries (cf. Bartley, 2007;McDonnell, King, & Soule, 2015)? How do corporations change their investment and implementation strategies in response to mobilization; does delayed or reduced expansion in one area mean increased expansion in another?
Finally, while throughout this special issue we emphasize getting a favourable policy on the agenda or getting a law passed as outcomes of organized collective action, a critically important remaining question concerns how movement coalitions and their opponents influence processes of policy implementation. The adoption of policies is likely to set off "reactive sequences" (Falleti & Mahoney, 2015, 222), whereby attempts to implement a policy lead to new sets of events. These events may include attempts by movement groups to shape the trajectory of implementation and to use openings created by policy and institutional change to advance new demands, strengthening shifts toward social and environmental protections. Reactive sequences may also involve reversals and backlashes, resulting in iterative processes that may lead to conflicts with dominant coalitions over the terms and process of implementation, changing the direction of policy and institutional change. Conflicts over, or the tortured process of, defining the reglamentos of socio-environmental laws in the region point to the significance of this issue. What is at stake in all of these questions is the larger issue of "alternative" development and what this idea means for different actors and publics, as well as how they envision attaining it.

Notes
1. The concept of extractivism is not new in Latin America. The term has been used to refer to historical uses of natural resources, such as the rubber boom of the late nineteenth and early twentieth centuries, and more recently, to the large-scale extraction of mineral and hydrocarbon resources. When referring to extractivism, we follow Gudynas (2013), who suggests that extractivism refers to forms of extraction that: involve the removal of very large amounts of soil, rock, vegetation and water in the process of accessing products of financial value; lead to large-scale and widespread landscape disturbance; involve the export of largely unprocessed raw materials; and involve global commodity chains. For some authors, extractivism also refers to a particular model of development that prioritizes the control and export of natural resources by large-scale enterprise as a basis for economic growth. Used this way, the term refers to a style of development. However, extractivism does not offer a theory of development as it does not account for how large parts of the economy or society work. Other concepts and theories are needed to explain why and how resource extraction becomes prominent in a society and the conditions under which large-scale control and export of natural resources becomes a central component in a national development strategy. Extractive industries are one form of mega-development. 2. Some opposition targets private corporations (cf. Walker et al. 2008;Bartley and Child 2014). At this stage, we are primarily interested in opposition that seeks change in government action and policy, though we consider how private sector actors fit into processes of organized collective action targeting state policies. 3. Shifting the venue to international organizations may not necessarily yield outcomes favourable to movements, since some international institutional settings may be less accessible and permeable to pressure (see Spalding, 2018). 4. This framework is grounded in existing approaches, but describes the types of effects differently. First, whereas Giugni (2004) and Giugni and Passy (1998) refer to "indirect" movement effects as effects that are mediated by either political allies or public opinion, we use "indirect" effects for situations where movements have no direct input into the policy formulation process. Second, while for Giugni, "joint" effects rely on pressure from both social movements and allies in institutional settings, we use "mediated" effects to describe cases where movements advance their agendas by working with political allies.