Converging Divergence: the Diffusion of the Green State in Latin America

Abstract

Global processes of policy diffusion result in different types of state development. A broad view of environmentalist reform in Latin America easily reads as top-down diffusion of blueprints and institutional convergence. But such a thesis is reductionist and ultimately misleading, case studies demonstrate. First, diffusion mechanisms matter for divergence: when normative and mimetic mechanisms are relatively strong vis-à-vis coercive forces, formal state change is followed by more meaningful real state change; when the coercive mechanism rules unmatched, green state change ends up being formal for the most part. Secondly, institutional entrepreneurs face shifting opportunity structures for political change; because these opportunities are never uniform, national experiences will differ. Thirdly, national institutional environments provide contrasting domestic resources and cultures for the building of green states; legacy, in short, will condition translation by entrepreneurs. A bridging institutionalist framework helps us make sense of "converging divergence".

St Comp Int Dev (2014) 49:242265

DOI 10.1007/s12116-013-9147-6

Jos Carlos Orihuela

Published online: 27 December 2013# Springer Science+Business Media New York 2013

Abstract Global processes of policy diffusion result in different types of state development. A broad view of environmentalist reform in Latin America easily reads as top-down diffusion of blueprints and institutional convergence. But such a thesis is reductionist and ultimately misleading, case studies demonstrate. First, diffusion mechanisms matter for divergence: when normative and mimetic mechanisms are relatively strong vis--vis coercive forces, formal state change is followed by more meaningful real state change; when the coercive mechanism rules unmatched, green state change ends up being formal for the most part. Secondly, institutional entrepreneurs face shifting opportunity structures for political change; because these opportunities are never uniform, national experiences will differ. Thirdly, national institutional environments provide contrasting domestic resources and cultures for the building of green states; legacy, in short, will condition translation by entrepreneurs. A bridging institutionalist framework helps us make sense of converging divergence.

Keywords Diffusion . Global blueprints . Green state . Institutional entrepreneurs . Institutionalisms

Introduction

During the last four decades, states around the globe have adopted ministries of the environment, trans-sectoral regulatory agencies, national park systems, air and water

This article is the ultimate outcome of a trans-disciplinary journey that started with my doctoral research, a quest to understand how institutions matter for development. I am grateful to my advisors Rosemary Thorp and John Coatsworth as well as to Lisa Anderson, Albert Fishlow, Ira Katznelson, Jos Antonio Ocampo and Maritza Paredes, and Diego Snchez-Ancochea, who read and commented my dissertation work at different stages. Fieldwork research was funded by the Institute of Latin American Studies at Columbia University. The translation and further transformation of the thesis into this manuscript benefitted from comments by Patrick Heller, Richard Snyder, Tatiana Andia, and Andrew Schrank. I am also indebted to SCID anonymous reviewers and editors. An earlier version was presented at LASA 2012. Emily Kirkland and Alejandra Ziga provided excellent research assistance.

J. C. Orihuela (*)

Departamento de Economa, Pontificia Universidad Catlica del Per, Av. Universitaria 1801, San Miguel( Lima 32, Perue-mail: [email protected]

Converging Divergence: the Diffusion of the Green State in Latin America

St Comp Int Dev (2014) 49:242265 243

quality standards, and environmental impact assessments. I label such state organizations and regulations green state structures. Latin American countries have also produced versions of such constructs, despite differences in income per-capita, industrialization, and political systems between Latin America and OECD countries and within the region (Braes 1991; Van Cott 1995; Acua 1999). If policymakers were simply implementing global blueprints, we would expect green states structures to be identical across Latin America. This is not the case. Instead, different countries have produced very different green states. There are shades of green, a phenomenon that Dryzek et al. (2003) documented for advanced capitalist economies.

This article argues that a top-down causal imagery to explain the diffusion of environmentalist organizations and regulations across states exposed by Frank et al. (2000), influenced by the institutionalism of John Meyer (Meyer et al. 1997), is not an accurate reading of facts. To make sense of institutional change, we need to understand better the underpinnings of the observable macro pattern of green state diffusion. To do so, I put forward a simplified framework for thinking about institutional change, one that bridges sociological and historical institutionalisms. The effort is aligned with the call Campbell and Pedersen (2001) and Campbell (2004) made for a second move of institutional analysis: to establish a more fluent dialogue between neo-institutionalist traditions in order to construct more nuanced analytic frameworks for the study of institutional change. Other noteworthy attempts of setting bridges among new institutionalisms include Pierson (2004), Greif (2006), and Hall (2010). I share the view that institutionalist traditions talk too little to each otherironically because of institutionalized incentives and belief systems that shape social scientific choice. By paying more attention to the insights of each traditionon the significance of bounded rationality, history, social environment, and ideaswe can offer more robust explanations of real-world phenomena and develop more accurate, nuanced theories. I contend that sociological institutionalism helps us think about the driving forces of diffusion and formal convergence, while historical institutionalism helps us understand why such forces have very different results in different places. Looking at the fine grain of the macro process of global diffusion of state organizations and policies, with case studies focused on the work of domestic institutional entrepreneurs and the actual mechanisms of innovation and translation hiding inside observed diffusion, I am able to demonstrate that there is more to state change than top-down homogenization of state structures.

The remainder of the article is organized as follows. I begin with a broad cross-country view of the formal diffusion of the green state in Latin America, showing the convergence of formal policy, regulations, and state organizations. I then present my bridging institutionalist framework, which will be used later in the article to analyze the cases. After the conceptual discussion, I turn to the case studies with the help of primary research carried out in Chile and Peru, green state latecomers, and secondhand analysis of the literature on Brazil, Latin Americas green state leader. The analysis highlights the diversity of diffusion mechanisms, opportunity structures, and institutional legacies in the three countries; these factors should lead us to expect diverging institutional change. The conclusion assesses the contributions and limitations of the discussed analytic frameworks, namely new institutionalism in sociology and historical institutionalism, and the need for an integrative approach to situate the analysis of policy choice.

244 St Comp Int Dev (2014) 49:242265

A Broad Look at State Change Shows Convergence

A broad look at state organizations and national environmental laws across Latin America shows a pattern of homogenization. As shown in Fig. 1, the process has had two stages, beginning in the early 1970s: first, the adoption of trans-sectoral regulatory/coordinating agencies, which nominally resemble the United States Environmental Protection Agency (USEPA); second, the adoption of ministries of the environment. This second phase has also brought new and more substantial legislation, characterized by national laws of environmental protection (regulating environmental rights of citizens) and norms of environmental impact assessments (regulating environmental obligations of large-scale private and state-run economic activities). The pattern found by Frank et.al. (2000) for the second half of the twentieth century has continued: in the twenty-first century, Latin American states are forging ad hoc environmentalist bureaucracies, also called national services of environmental protection. Latin American policymakers hope to produce regulatory systems that are more autonomous than ministries of the environment, which are seen as permeable to day-today politics.

A narrative of top-down policy diffusion could be easily written from this evidence. Global environmental governance is a key part of this narrative. The 1972 United Nations Conference on the Human Environment in Stockholm fits as the opening salvo for the diffusion of green state blueprints. Once the United Nations Environmental Program (UNEP) was created in Stockholm, problems previously seen as local or national (or problems that were simply unseen) were defined as transnational in mainstream scientific and policy discourse. These problems included the production and trade of hazardous chemicals and toxic waste, deforestation, biodiversity loss, desertification, and environmental refugees, to name the most prominent. Scholars revisited Malthusian questions about the sustainability of growth (Meadows et al. 1972; Conca and Dabelko 2010). They challenged the contemporary economic paradigm on

20

18

16

14

12

10

8

6

4

2

0 1972 1977 1982 1987 1992 1997 2002 2007

Trans-sectoral / coordinating agencies Ministries of the environment National environmental laws

Fig. 1 The diffusion of trans-sectoral agencies, ministries of the environment, and National Laws of the Environment in Latin America, 19722012

St Comp Int Dev (2014) 49:242265 245

two fronts: the environmental impacts of and the environmental constraints to capitalism. The new policy paradigm that emerged is now commonly referred to as sustainable development after the Brundtland Commission report. The hope was that economic progress and environmental well-being could be reconciled if the right institutions were in place. In the emerging consensus of the global center, states had to green their political economy if the global village was to have a common future (World Commission on Environment and Development 1987). It would not take long for this wave of new thinking to penetrate the global periphery.

A closer look at the transnational activism that accompanied Stockholm might lead us to question the top-down narrative of green state diffusion. As Keck and Sikkink (1998) and the transnational networks literature has shown (Tarrow 2005), the process by which norms become universal neither starts nor ends with international governance and formal nation-state structures, and the networks of scientists holding policy power. Civil activism has to be brought in to understand the progressive diffusion of cultural values and legal norms around the globe. As in Polanyi (1944)s classic study of political economy, society embeds state environmentalisms, and new-era global environmental governance (Haas 1992). Historical facts better fit this second, more nuanced account. In particular, momentum for Stockholm 1972, UNEP and new forms of state action was heavily influenced by civic activismboth domestic and transnational. Post-war social movements had produced multiple 1968s, reinforcing a new imagery of goals and means for states and societies: a clean environment was a human right and political participation did not finish with the election of authorities. States responded to societies. In 1969, the US passed the National Environmental Policy Act; 1 year later, the Environmental Protection Agency was created. In short, environmentalism existed in the grassroots before it reached formal institutions in the top, and grassroots activism shaped the very production of new global blueprints (McCormick 1989; Rootes 1999; Dryzek et al. 2003).

How about the then non-heavily industrialized global periphery, say Brazil? At Stockholm 1972, Brazil was one of the delegations that eloquently expressed its reservations with the emerging environmental protection paradigm: developing countries had the right to make decisions about their natural resources, just as developed countries had. That meant clearing forests, depleting fisheries, polluting cities, and so on (Guimares 1991; Miller 2007). Yet Brazil, run then by the modernizing and developmentalist military, ended up taking the lead in the crafting of green state structures in Latin America. Moreover, the nation-state of Brazil exercised its geopolitical influence in the very making of the rules of the game for the postwar era. Thus, the next milestone of green global governance, the United Nations Conference on Development and the Environment took place in Rio de Janeiro in 1992. The conference was called Development and the Environment, as opposed to Human Environment, in accordance with the wishes of developing nations. In Rio, government officials and appointed national experts from the periphery participated in the production of globalized discourse and regulations (Guimares 1991; Miller 2007; Hochstetler and Keck 2007). Rio 1992 became a milestone for the diffusion of the ministry of the environment blueprint (see Table 2 in the Appendix).

Among middle-income Latin American economies, Brazil has been a leader in the production of formal green state structures, while Chile and Peru constitute late followers. The big regional players, Mexico (in 1972), Argentina (in 1973), and

246 St Comp Int Dev (2014) 49:242265

Brazil (in 1973) established state organizations that proposed and coordinated new norms and standards with regards to pollution, environmental conservation, and the rational use of natural resources. In Brazil, this diffused organizational model was called Special Secretary of the Environment, SEMA, modeled on the US Environmental Protection Agency (McAllister 2008, 22). But most of the legal and institutional developments that came along with SEMA and its equivalents were ultimately more theoretical than real (Guimares 1991; McAllister 2008). In 1974, Colombia passed the first comprehensive environmental law, the National Code of Renewable Natural Resources and the Protection of the Environment, which became a model to environmental lawyers elsewhere the region.1 In 1977, Venezuela set up the first Ministry of the Environment, a choice that was not to be followed by other Latin American countries until a decade later (Acua 1999; Braes 1991).

Formal institutional change waited two decades to reach Chile and Peru. Since the 1970s, these two countries had seen plenty of piecemeal evolution and aborted attempts at big-bang reform: examples included the air quality standards set by the Ministry of Health in Chile and the Natural Resources and the Environment Code in Peru, which lasted less than a year. But a trans-sectoral regulatory body and a national law of the environment as such had to wait until the 1990s. In 1981, Brazil created a normative and coordinating agency, the National Commission of the Environment (CONAMA); this meant integrating SEMA as an executive arm of CONAMA and reorganizing other rapidly evolving green state structures (Hochstetler and Keck 2007). By 1994, the brand CONAMA had travelled to Chile and Peru, but it would translate into quite different state organizations, as I will show two sections below. By 2010, CONAMAs of Chile and Peru (called CONAM) had mutated to become Ministries of the Environment. Brazil established its first version of the ministry blueprint in 1985.

How a Bridging Institutionalist Framework Helps Us Understand Green State Change

I will discuss the development of green states in Brazil, Chile and Peru in much greater detail later in the article, illustrating the role of converging divergence in these three countries.2 In order to do so, I present first an institutionalist framework that bridges sociological and historical institutionalisms for the study of state change, advanced in Author 2013a. I find that new institutionalism in organizational sociology provides a usable framework for understanding diffusion and forces favoring homogenization, while historical institutionalism helps illuminate idiosyncratic state change. This is not to deny that insights about path-dependence can be gained from organization sociology literature or about diffusion from historically oriented social science research. John L. Campbell (2004) gives the best exposition on the past and future of cross-pollination I am aware of. Still, a key message in one literature is that the embedding social environment matters and produces institutional diffusion, at the same time that a key message in the other literature is that history matters and produces institutional path-dependence. One institutionalism makes us think more directly about convergence; the

1 Interview with Manuel Pulgar Vidal, former executive director of the Peruvian Society for Environmental Law and current environment minister of Peru, September 27, 2007.

2 The term converging divergence has been borrowed from a manifesto for reinvigorating the political economy of Latin America made at LASA 2013.

St Comp Int Dev (2014) 49:242265 247

other, about divergence. A bridging institutionalist perspective, then, helps us make sense of converging divergence.

Institutional Diffusion

The homogenization of institutional structures has long called the attention of social scientists. Revisiting Webers thesis on bureaucratization, sociological neoinstitutionalism argues that organizational convergence results from the structuring of organizational fields, carried out largely by the modern state and professions (Di Maggio and Powell 1983; Giddens 1979). Indeed, the case of the environment fits well into this general theory. In the post-war era, modern notions of wilderness and welfare were reconfigured with help of new-era professionals such as scientists and lawyers of the environment (Meyer et al. 1997). Soon, professional and policy leaders had adopted new ideas about nature and society. Environmental regulation became a natural ingredient of nation-state governance, joining national security, welfare, economic growth, fiscal and monetary responsibility, and basic health and education for all.

How does homogenization take place? How does a USEPA or a Ministry of the Environment become a global blueprint? The organizational school of institutional analysis contends that adoption is largely driven by a search for legitimacy in a socially constructed environment, a world society, rather than by purely rational decision making driven by competitive forces (Di Maggio and Powell 1983; Meyer and Rowan 1977). Agents and organizations respond to their environments, which are composed of other agents and organizations playing in the same evolving field (Di Maggio and Powell 1983; Schelling 1978). Outside the marketplace, where the competitive mechanism rules, homogenization is driven by institutional mechanisms. New institutionalism in organizational sociology classifies these mechanisms as coercive (resulting from the need for political influence and legitimacy), mimetic (resulting from standard responses to uncertainty), and normative (resulting from professionalization) (Di Maggio and Powell 1983; Dobbin et al. 2007; Strang and Meyer 1993).

The early views on the homogenizing role of institutional environments on organizations have been revisited. There is now a more nuanced understanding of isomorphism by scholars who pay attention to the interpretation, manipulation, revision, and elaboration performed by the affected actorsthe bureaucrats and policymakers who put new ideas and models into practice (Scott 2008, 430). In fact, the founding article of Meyer and Rowan (1977) exposes the mechanism of decoupling, by which organizations forced to adopt new models by their institutional environment end up taking into account local circumstances and practical realities, and so produce actual practices that do not match the imported blueprints. A branch of the organizational sociology literature then finds superficial conformity behind isomorphism (Meyer et al. 1997; Scott 2008). A much less ambiguous enunciation of policy diffusion is to call it translation, which finds the author/recipient dichotomy useless (Bockman and Eyal 2002). Agency is more prominent under the concept of translation, in which involvedrather than affectedactors use local principles and practices to adapt spreading foreign principles and practices (Hironaka et al. 2002; Campbell 2004). Global diffusion comes along with national innovation, and so may well produce diverging outcomes (Chorev 2012; Ban 2013; Gilardi 2012). Actor-network theory, from a different tradition, offers an agentic view of translation, which would say that what

248 St Comp Int Dev (2014) 49:242265

this article calls institutional entrepreneurs become so because of their success in opening flowing black-boxes, i.e., green state blueprints, and strategically reinventing their meaning (Latour 1987; Callon et al. 1981). If agents are able to translate and reinvent models imported from abroad, then top-down institutionalist accounts such as Frank et al. (2000) do not fit reality well.

Institutional Path Dependence

While new institutionalism in organizational sociology has focused on institutional diffusion and homogenization, historical institutionalism has revisited path-dependent institutional change. Across disciplines, scholars attempt to clarify the meaning of history matters (Mahoney 2000; Pierson 2004; Thelen 2003). As scholars revise their thinking about big structures and large processes, institutional development through policymaking receives new attention (Pierson 2006; 2004). The interest in critical junctures as windows of opportunity for radical institutional reform is now complemented by a refreshing focus on the everyday forms of institutional change (Mahoney et al. 2010; Streeck and Thelen 2005; Taylor 2009). Timing matters a great deal for institutional change: this article shows that policy entrepreneurs create opportunities from both once in a life time events (e.g., Stockholm 1972 and Rio 1992, return to democracy, and Free Trade Agreement junctures) and everyday contingencies. Both big-bang and piecemeal bureaucratic change can have lasting institutional consequences; for instance, I will argue later that structural adjustment shaped the birth of a weak green state of Peru, while long-run everyday policy leadership created the culture of technocratic autonomy that gives form to contemporary green statecraft in Chile. Moreover, since the social realm is characterized by multiple equilibria (i.e., the same origin leads to a range of possible ends depending on interacting social dynamics) (Hoff and Stiglitz 2001), the tracing of historical processes is central for the insightful understanding of (path-dependent) institutional evolution (Pierson 2004). The existence of various institutional types should not be a surprise, but we will need to look seriously at history in order to understand why a given type developed in a particular location (Tilly 1995).

The cultural turn of the social sciences, which influenced most emerging neoinstitutionalisms (Hall and Taylor 1996; Greif 2006; Scott 2008), helps to explain the contentions of historical institutionalism about the shaping properties of history. Over time, policy behavior creates state resources and belief-systems or mental maps, which feed the self-enforcing properties of institutions. In addition to developing organizational resources (regulatory systems, non-regulatory executive capacities, human capital, know-how, and information systems), which matter a great deal for building on new state action (Sikkink 1991), states can produce culture (normative and cultural-cognitive frameworks). Thus, policymakers build on but also are shaped by old state action, usually through interacting cognitive (culture and ways of learning), returns-to-scale (strategic use of available institutional resources), and feedback (pass and present state action feeds new state action) mechanisms (Pierson 2004).

Some forms of state building become natural for state crafters; others, unthinkable. Institutions construct actors. Although this should not be read in a deterministic sensecausality also goes in the opposite direction (Giddens 1979)agency is embedded in its historically built institutional context. Insights from historical

St Comp Int Dev (2014) 49:242265 249

institutionalism and broader institutional theory suggest that national politicaleconomy systems, which I call institutional environments in the case studies section, are likely to produce green states resembling pre-existing state structures, as in a somewhat radical version of the varieties of capitalism school (Hall and Soskice 2001): Brazil, Chile, and Peru all produced a National Commission for the Environment, a CONAMA or a CONAM, but the national CONAMAs should not be expected to be identical twins, much as other state organizations with equivalent functions are not identical across the three countries. Before the era of environmental regulation, the developmental state paradigm took diverse national formsled by the BNDES in Brazil, CORFO in Chile, and the late and short-lived National Institute of Planning in Peru. Thus, even within Latin America, there were varieties of developmental statesjust as there are now different varieties of green states (Thorp 1998; Ros and Ocampo 2011). Given the play of economic interests and policy passions, the history of state-society relationships is fundamental for the molding and re-engineering of state forma and action (Kohli 2004; Dryzek et al. 2003; Migdal 2001).

The rise of the green state is a global phenomenon that illustrates the process of institutional diffusion theorized by sociological institutionalism; yet, as historical institutionalism would have predicted, the resulting picture of state-run environmentalism shows heterogeneity behind homogenization, to be explained in detail in the coming section. Convergence and divergence end up being two sides of the same process: after green state diffusion, states are more alike but still quite particular. Similarities among green states exist, but mostly on paper. Isomorphism is little more than formal largely because the opportunities, resources, and belief systems shaping arenas for policy agency are not uniform across nations. Summing up, the integration of the frameworks depicted here makes us expect converging divergence as a general property of institutional change.

Case Studies Show Divergence: Diffusion, Entrepreneurs, and Translation

Frank et al. (2000) does not provide an accurate account of the construction of the green state. It artificially opposes a top-down global explanation with the bottom-up domestic alternative to say that the former fits reality better than the latter, while saying nothing about the quality of state constructs that result from diffusion. I will come back to the latter below when developing the cases. On the former, the transnationalist view of Keck and Sikkink (1998) gives a more nuanced framework to discuss the general trend and particular cases of environmentalization. In short, there are more than two mechanisms at work (the top-down vis--vis the bottom-up) and more than two self-contained arenas where the action takes place (the global vis-vis the domestic); it is in the study of the co-production and interplay of mechanisms and dimensions that scholars gain insight. Thus, the global dimension usually works as a key contextual variable that recreates opportunity structure for fundamentally locally driven processes, drawing on the contentious politics literature (Tarrow 1998; McAdam, Tarrow and Tilly 2001); and domestic processes are never fully autarchic, as Hochstetler and Keck (2007) discuss richly for the greening of Brazil. The domestic is shaped by and shapes the global, as both arenas are inhabited by actors embedded in both domestic and global social networks. Rooted cosmopolitans are to be found in states and societies (Tarrow 2005), as are globalized nationalists.

250 St Comp Int Dev (2014) 49:242265

By focusing on the real-world agents who imagine, call for, copy-and-paste, translate, and carry out policy and institutional change, one is able to escape the false dilemma of choosing between artificially rival explanations and, instead, to look for a more nuanced understanding of diffusion phenomena. The institutional entrepreneurs responsible for crafting the nationalist will to improve (c.f. Li 2007) are part of domestic development processes while also sharing, to an extent, their normative and cognitive frameworks, with peers around the globe. It follows that, in principle, blueprints of what the nation-state must be and should do can be co-produced in different sites, or can originate in laboratories of the periphery, as shown in the insightful account of the Eastern European roots of neoliberalism by Bockman and Eyal (2002). In his classic study of how reform-mongers fought for structural change in Latin America, Albert Hirschman (1963) showed that foreign expertise was frequently brought in to legitimate policy agendas advanced by domestic policy agents. Thus, to say that what to do with economic development was dictated in a global production site (the center in Latin American structuralism) and performed in a receptor site (the periphery) over-simplifies reality.

Through close inspection of institutional entrepreneurs, scholars can scrutinize the mechanics of diffusion and thereby qualify the general picture of convergence. Scholars who believe that history and domestic institutional environments matter will not be surprised to learn that organizational models and legal blueprints that diffuse around the globe tend to be implemented differently in each country. The contribution of this study is to give a micro-level account of how such a claimed general property of institutional changeas seen defended in some but not all institutionalist quartershas worked in the case of green state diffusion in Latin America. Looking at how and why the agents of green state change carried out reforms the way they did, I found that different diffusion mechanisms produce different types of state changes, that bounded institutional entrepreneurs strategically use opportunity and legacy to build on institutional change, and that as a result institutional entrepreneursby means of innovation and translation end up producing unique domestic varieties of globally diffused models of state action.

The Variety of Diffusion Mechanisms Leads Us to Expect Divergence

Diffusion mechanisms are analytically distinguishable but empirically blurry. With this call for sympathy towards nuanced argumentation, I turn to the evidence beginning with the leader of the trio, Brazil. Why would the authoritarian and development-minded Brazilian state end up heading the formal environmentalist journey in Latin America? Pure competition is not the answer, since Brazils favorite rival, Argentina, left the race 3 years after starting it, when its rising dictatorship aborted newborn state environmentalism. Coercion does not explain the whole story either. Brazil has a big, industrialized domestic economy, in addition to a large territory rich in natural resources, with domain over the Amazon basin and the South Atlantic. Then, it has much more freedom when it comes to negotiating in the international arena than any other Latin American state. The mimetic mechanism played a more evident role, especially given the uncertain circumstances, which included changes to the global understanding of progress and development of nations. Created in 1973, the emerging SEMA was modeled on what state-makers believed to be a worlds best-practice for environmental regulation, USEPA (Drummond and Barros-Platiau 2006; McAllister 2008).

St Comp Int Dev (2014) 49:242265 251

The normative mechanism is apparent too: a critical mass of state officials had gained green consciousness. This is not to say that the entire Brazilian government took an environmentalist standpoint. Thus, the participation of SEMA in the key development schemes of the 1970s, such as the Nuclear Program, the Carajs Iron Mine Project, the Pro-Alcohol Program, the Itaipu Dam, and the Transamazon Highway, was limited to a rubber stamp review prerogative (Guimares 1991). As discussed in the subsection below, protected areas were not expanded in Brazil until the late 1970s, long after the creation of SEMA; clearly, the greening of the state was a piecemeal rather than a big-bang institutional process. Thus Brazil was a leader in the formal establishment of regulations and trans-sectoral state organizations (beginning in the early 1970s), but that does not mean that Brazil was a leader when it came to actual environmental protection.

In green laggard Chile, the same mechanisms of institutional diffusion were present, but in different doses and, crucially, at a different time. The case of Chile helps to highlight the importance of political supportor lack of uncompromising opposition for institutional entrepreneurial activism to be effective. The story of green state change starts much more slowly in Chile because the Pinochet regime (19731990) was radical in its friendliness towards markets. Under Pinochet, new state institutions were out of the question, and green state activists could not hope for feedback effects or support from other branches of the state. Still, there was plenty of formal piecemeal change, from the 1978 first National Air Quality Standards (never enforced) to the 1984 National System of Protected Wild Areas Law (Camus 2006; OECD 2005). A proper body of national environmental law and a trans-sectorial state organization had to wait until the dictatorship was gone, when Brazil and most Latin American states were already experimenting with ministries of the environment. After Pinochet, moreover, the institutional legacy of the free-market doctrine meant that even the left-of-center Concertacin regime (19902010) did not make any radical departures from the basic paradigm that favored market mechanisms over governmental regulation (Katz and Del Favero 1993; Silva 1996; Carruthers 2001).

Coercion and competition were as much present as norms and mimesis, and catalyzed the 1990s reform process. The forthcoming negotiation of a Free Trade Agreement (FTA) with the US shaped and empowered environmentalist reform discourse, especially coming right after environmental boycotting that almost endangered the North Atlantic Free Trade AgreementMexico had to prove it was committed to environmental regulation before the US would sign. Then, the FTA made improving environmental institutions a must, creating a pragmatic rationale that allowed the environmentalists within Concertacinled by lawyers and mining engineersto find a middle-ground with the economists and the pro-business policy establishment. Professionals in Chile were already trying to implement green state structures, driven by their perceptions of global norms; the FTA added competitive and coercive pressure to this process (Carruthers 2001; Lacunza 1996; Orihuela 2014).

In environmental laggard Peru, one again finds all three institutional mechanisms of diffusion at play, but the relative importance of the mechanisms and the timing of diffusion are once more case-specific. In the 1970s, relatively new and cohesive networks of professionals, mostly forest engineers, made their way into the state apparatus. Radical or moderate, the ruling military (19681980) favored state planning, which enabled institutional entrepreneurs; forest engineers, and biologists transformed

252 St Comp Int Dev (2014) 49:242265

Peru into an active conservationist state (Dourojeanni 2009). State environmentalism weakened with 1980s democratization: elected state authorities were not sympathetic to the green cause; Congress, in turn, did not pass the Colombia-inspired Natural Resources and the Environment Code (Dourojeanni 2009; Pulgar-Vidal 2006). By 1994, however, international development forces had coerced Peru to create a Chile-inspired CONAM. In particular, transnational capital in the extractive sectors, returning to Latin America after the Lost Decade, demanded environmental permits and regulatory blueprints to begin business (Orihuela 2014).

The variety of institutional mechanisms of diffusion leads us to expect divergence. When the coercive mechanism rules, institutional change is unlikely to be as vigorous as when normative triggers have equivalent importance. This can be learned from comparing institutional change within Peru: 1990s CONAM vis--vis 1970s forest protection. The same can be found when looking at the adoption of the CONAMA model from Brazil by Chile (lower relative importance of coercion and competition) and Peru (higher relative importance of coercion and competition). The national CONAMAs and the ministries of the environment discussed below end up having in common little more than the label and a set of declarative mandates.

Institutional Entrepreneurs Need Opportunity and Are Not Ultra-agentic Agents

As seen in the three cases, the in-depth inspection of state change highlights the significance of opportunity. Historical contingency situates agency, which is another reason to expect that internationally diffused ideas and models will unfold differently in different places. The cases have already shown that in all countries, at all times, there are policymakers driven by global norms who are forging piecemeal change and radical reform; yet their significance for state craft is not always the same. Coercive and competitive forces, in particular, can add pressure and urgency, as when the international community in the 1990s told Chile and Peru that to trade and get foreign capital they had to go green. That is the global timing, but there is also the local one. By 1990, Chile was moving from a neoliberal dictatorship to a left-of-center democracy with sympathy towards (market-friendly) regulation, while Peru embarked into a wholly hearted neoliberal experiment skeptical of any form of state interventionism. Thus, domestic forces can constrain norm entrepreneurs and give a different meaning to coercion and competition. In particular, it becomes more difficult to create an autonomous, professional environmental bureaucracy under a radical domestic free market ideology. There will always be private economic interests that oppose new state making, to be sure, but public political passions matter too. Contingency interplays with diffusion mechanisms, enhancing some (e.g., the coercive and competitive mechanisms in both Chile and Peru) and constraining others (e.g., the normative mechanism in Peru). In short, opportunity qualifies diffusion.

First SEMA chief Paulo Nogueira Neto does not fit the metaphor of an enactor of global blueprints. Instead, he was an innovator and a translator, an agentic actor co-producing global blueprints. A lawyer and a biologist, Nogueira Neto was since the mid-1950s part of a small yet significant domestic network of activists. He founded one of the first Brazilian conservationist associations and participated in the international conservationist movement. As chief of SEMA, he built on the social legitimacy that accompanied his professional status as a scientist. The argument for the case of 1970s

St Comp Int Dev (2014) 49:242265 253

Brazil is that normative triggers were strong because a handful of creative environ-mentalists managed to become prominent-though-constrained within government. The reach of these agents would never be as strong as to stop state-fostered industrialization and forest colonization, but plenty of piecemeal progress was achieved thanks to their activism that went beyond the reception of blueprints (Drummond et al. 2009; Hochstetler and Keck 2007).

Innovation and translation were two mechanisms operating in everyday bureaucratic life. Nogueira Neto called the state-making tactics he employed environmental guerrilla activities: supported by a network of green-minded Brazilians, he tried to occupy available bureaucratic territory, making good use of small but visible interventions. But SEMA and its chief Neto were not alone in the guerrilla warfare within the state. When the SEMA expansion began in 1978, the Brazilian Institute for Forest Development (IBDF), the first fully fledged Brazilian agency dedicated to the management of renewable resources, had only approved three parks and one biological preserve in its 11-year history. In contrast, the same organization, using the same legal framework, passed 20 new conservation units between 1979 and 1986, in addition to creating new managerial tools and regulatory schemes. What had changed was leadership, that of new chief Maria Teresa Jorge Pdua, who led the efforts of a small group of IBDF bureaucrats (Drummond and Barros-Platiau 2006).

In 1990s Chile, the key institutional entrepreneur belonged to a different community of professionals. Rafael Asenjo was part of a previously unseen generation of lawyers litigating environmental protection cases based on civil rights declared in the 1980 Constitution. That the judiciary system was autonomous enough despite the dictatorship enabled their form of activism. He became a public figure with his work to defend a fishing town from water contamination by state-owned copper-giant CODELCO. Within the first weeks in office, President Aylwin (19901994) made him the head of the embryo of what became the Chilean CONAMA. Weakened by the process of approving the national environmental law, passed by Congress only days before Aylwin finished his term, Asenjo abandoned government and became a critical supporter of the Concertacin under the administration of President Frei (19942000), who constrained in practice the formal bureaucratic autonomy of the new CONAMA (Orihuela 2014).3

Another key actor appeared at the critical juncture of the return to democracy: starting in 1991, mining engineer Jaime Solari led the regulation of air pollution from state-owned smelters and eventually became a fundamental broker for the passing of the 1994 environmental legislation.4 He headed a small environmental unit within the Ministry of Mining, building on the work of the network of engineers that had put mining pollution onto the policy agenda. Before joining government, Solaris professional experience as an environmental engineer included Brazil, and so he knew firsthand how behind the Chilean state was in relation to what he thought to be a regional benchmark. He advocated within government and political elites that it was shameful that Chile did not have environmental institutions by 1990.5 In conjunction with the return to democracy, air pollution in capital city Santiago (worsening over the last few

3 Interview with Rafael Asenjo, head of the Secretara Tcnica de la Comisin Nacional de Medio Ambiente (which became CONAMA) in 19901994. Santiago, July 19, 2007.

4 Interviews with Gustavo Lagos, adviser to the Environmental Unit of the Ministry of Mining in 19911994. July 23, 2007, and with Jaime Solari. Santiago, August 24, 2007.

5 Interview with Solari.

254 St Comp Int Dev (2014) 49:242265

winters) had created an opportunity for championing the environment (Orihuela 2014). The rise of CONAMA thus resembles that of Brazils SEMA in 1973, when an air pollution event caused by pulp mills in Porto Alegre, the hometown of the cabinet chief, catalyzed the approval of the state organization (Guimares 1991).

In the case of Peru, the key institutional entrepreneur in the 1970s chapter of green state making was Marc Dourojeanni. He studied in Belgium and had a doctorate at hand. He was picked by the FAO as its counterpart in the military-run state for its protected areas initiativePeru being a potential poster child of biodiversity. Shortly afterwards, Dourojeanni joined the International Union for Conservation of Nature (IUCN), the meca of global conservationism in his view. He believes that his influence was enhanced by the professionalism of the military chiefs running the state.

6 By 1973, when he became General Director of Forestry and Fauna in the Ministry of Agriculture, a civil critical mass had formed, promoted first by the National Committee for the Protection of the Environment, which was founded in 1940 by intellectuals, geographers, and new-era environmental professionals. As part of the movement, the molineros, the agriculture and forestry engineers trained at the Universidad Nacional Agraria La Molina (UNALM), which opened its forestry department in 1963, progressively made their way into the state (Dourojeanni 2009). A close associate was Antonio Brack, who succeeded in creating a conservation area for vicuas, which seemed to be destined to survive only in the national seal. Three decades later, in 2008, Brack became the first minister of the environment in the history of the Peruvian state.

Green laggard Peru, a receptor site, would become a producing site of institutional innovations, and diffusion went from the periphery to the periphery. For these high-skilled activists turned statesmen, the regional example was Venezuela, not Brazil, a bureaucracy with which they had little contact. In fact, Peruvian state conservationists innovated policies that were years later adopted in Brazil. In particular, the Brazilian extractivist reserve is a version of the communal reserve, a Peruvian institutional innovation. The communal reserve was legally defined as a protected area where economic activities of existing communities were legal, given that conservationism in the Amazon had to face the reality of indigenous communities. IBDF Chief Jorge Pdua learned about the scheme from Dourojeanni himself, to whom she had married.7

The opportunity structure for the greening of Peru changed with the return to democracy. Throughout the 1980s, a handful of Congressmen acting as individuals from virtually every political party fought for approving a Natural Resources and the Environment Code, thereby giving practical meaning to the environmental rights of citizens declared in the 1979 Constitution. However, the environmental activism of Congressmen went nowhere as political chiefs were not persuaded to invest political capital in environmental protection, given the opposition of extractive industries like mining and fishing. The crusade for a national environmental law was supported by the civil environmental movement, particularly by a vanguard of young lawyers who ended up creating an influential NGO, the Peruvian Society of Environmental Law (SPDA).8 The key institutional entrepreneur of SPDA is Jorge Caillaux, believed to be the first Peruvian lawyer to write a dissertation on environmental law, advised by forest scientist

6 Interview with Marc Dourojeanni.

7 Ibid.

8 Ibid.

St Comp Int Dev (2014) 49:242265 255

Marc Dourojeanni himself. That the network of lawyers has built a relationship with the network of forest conservationists, and the two with the network of supporters of indigenous peoples rights and socio-environmentalism, gives a sense for the complexity and the cross-cutting interests of the civil activism that embeds institutional entrepreneurship (Orihuela 2013).

As can be expected in extremely hectic politics of hyperinflation, opportunity structure changed feverishly. But neoliberal ideology ended up triumphing and delineating an unfriendly arena for environmental regulation. In 1990, within the first weeks of the Alberto Fujimori regime (19902000)who stopped hyperinflation and went to emulate Pinochetian economic reformsthe Code came to life. The alignment of forces in the council of ministers changed within months, so that pro-investment legislation of 1991 made the 1990 Code toothless. To surprise of the ruling neoliberal technocracy, however, the international development establishment was keen to foster some form of state environmental interventionism. As part of structural reform conditionality, the World Bank and the Inter-American Development Bank funded programs of institutional strengthening and environmental reform in the Peruvian state. In particular, institutional entrepreneurs of the Chilean CONAMA were brought in with IADB funding to advise the design of the Peruvian CONAM (Pulgar-Vidal 2006, Orihuela 2014).9

Under such a problematic window of opportunity, no institutional entrepreneur could have the profile of Neto, Asenjo, or Dourojeanni. The pick for a chief of CONAM was not a champion of the environment. A businessman who had been helping government with privatization, Gonzalo Galdos, received the post. Once in office, he did not take a guerilla approach to gaining new governmental functions because he did not feel strongly about environmental regulation. He did not believe that, say, a CONAM rather than a Ministry of Energy and Mining should give the environmental permits to mining companies, as had been agreed in Chile. For Galdos, markets should regulate themselves and the role of CONAM was above all educational. So private-minded was CONAM in practice that in operated for 2 years exclusively with IADB funding in the headquarters of Galdos metal-mechanic enterprise, and Galdos never charged salary.10 Principal-agent theory proved valid even in such a particular policy environment. CONAM became a lobbyist within the state, a low-profile think-tank made of environmental lawyers supporting legislative and sectoral initiatives, but with minimal executive capacity (Pulgar-Vidal 2006).

Path-Dependence Helps to Explain Translation, Therefore Divergence

Along with diffusion mechanisms and the opportunity structure within which institutional entrepreneurs craft change, institutional legacy is a third reason to problematize the global diffusion of institutional models and to expect a pattern of converging divergence. The formal designs and policy solutions copied and adopted by green state entrepreneurs were framed by pre-existing domestic state-society traditions and state capacities. Thus, states can all be shifting in the same direction but each of them does so in highly idiosyncratic ways. The claim is that internationally diffused regulatory

9 Interview with Mariano Castro.

10 Interview with Gonzalo Galdos. CONAMs first President. Lima, January 11, 2010.

256 St Comp Int Dev (2014) 49:242265

blueprints (e.g., the national environmental codes and environmental impact assessments), organizational models (e.g., the transsectoral state agency and the ministry of the environment), and policy principles (e.g., enforcement of air quality standards and promotion of forest conservation/sustainable forestry) end up translated by local institutional entrepreneurs to resemble the kind of social construct that characterizes their domestic institutional environment. In other words, there is a mimetic mechanism working for global diffusion (therefore homogenization and convergence), but there is also a mimetic mechanism favoring domestic alteration (therefore path-dependence and divergence). The social environment that embeds reformist agency is composed by more than one institutional dimension; in particular, the national state-society tradition can be as important as the transnational epistemic community; therefore, at least two homogenizing forces (a transnational and a domestic) interplay pulling institutional change toward different poles.

Translation rules, but it is not unbounded. Diffused legal, organizational and policy models of green state action unfolded into particular national constructs. Thus, SEMA is supposed to have been modeled after USEPA, but the organizational structures and the bureaucratic powers established for SEMA by the Brazilian state were not equivalent to those granted to EPA in the USA. The USEPA integrated, in one agency, pre-existing state bodies from the Departments of Health, Education, Welfare, Interior, and Agriculture (Andrews 2006). The Brazilian state did not pass legislation and organizational resources of equivalent importance partly because it could not have done so. EPA in the US state was given much more bureaucratic autonomy than SEMA, which was established within the Ministry of Interior of the Brazilian state (Andrews 2006; Guimares 1991).

Established SEMA, subsequent green state-craft in 1970s Brazil is a rich illustration of how institutional legacy shapes but does not determine policy agency and further institutional production. The reformism of SEMA chief Neto was both constrained and enhanced by pre-dating state rules, traditions, and organizations, as well as state-society relations. Neto reinvented SEMAs mandate and organizational structure, allowing it to take over pre-dating conservationist functions: given that national parks were in hands of another state organization, he came out with a proposal for so-called ecological stations dedicated to apparently innocuous scientific purposes. The approved stations, however, were entitled by law to hold up to 10 % of forestland for research. Through piecemeal policymaking, then, Neto and team reinvented the geography of national parks: between 1977 and 1986, SEMA created 38 ecological stations, 11 environmental protection areas, and a small set of other conservationist schemes (Drummond and Barros-Platiau 2006; Hochstetler and Keck 2007).

This wave of state conservationism led by Nogueira Neto can be better thought of as a third layer in the process of modern green state building in Brazil. The first layer has as milestones nineteenth century experiences of strong symbolic importance above all, the establishment of Tijuca Forest in 1844, long before the creation of Yellowstone, to become a global model. The second layer refers to the early twentieth century wave of national regulations and state organizations, which includes the Forest Service (created in 1921) and the first Forest Code of 1934. The second generation of conservationists behind these measures thought of forest protection as part of the modernization project of nation-state building (Drummond et al. 2009). In turn, third generation state conservationists of the late twentieth century were influenced by the

St Comp Int Dev (2014) 49:242265 257

emerging sustainable development doctrine. Three corresponding layers can be found in the history of environmental regulation in the USA, with USEPA embodying the third layer (Klyza and Sousa 2008). In each phase, Brazilian institutional entrepreneurs were influenced, but not entirely governed by spreading global blueprints, since each period of green state layering also had a powerful vector of domestic processes behind it (Drummond 1996; Miller 2007; Drummond et al. 2009).

In Brazil, decades-old state positivism offered unintended institutional resources for the green cause. Brazilian bureaucratic politics integrates a well-rooted tradition of patrimonial practices with a long-nurtured tradition of positivist bureaucratization (Drummond and Barros-Platiau 2006; Evans 1995). The techno-bureaucracy provided newborn SEMA with operating state structures to build on, as is clear from the highly technical quality of the Brazilian delegation to Stockholm 1972 and the various attempts of chief military statesmen to promote environmental policy before the birth of SEMA (Guimares 1991). On the other hand, the wild juxtaposition of state organizations (federal, regional, and local) with the patrimonial network of public officials in charge of conservation and environmental health made neo-positivist green state crafting difficult (McAllister 2008). Counting the federal, the state, and the local levels, SEMA had direct or indirect relationships with about 300 state organizations (Guimares 1991).

The legacy of participatory politics also shaped the rising Brazilian green state, including the production of CONAMA, a new black-box for green state making in Latin America. The greening of Brazil began with policymaking at the local level in the late 1960s, years before Stockholm and SEMA. Urban planners were key actors here. The significance of urban planners for local government during the dictatorship and later can be seen in the translation of the ministry blueprint into a Ministry of Urban Development and the Environment. The city of Curitiba, in particular, would become a global paradigm for urban planning and city greening (a producing site of a global script), thanks to a vanguard of urban planners that gave continuity to city management, which also incorporated citizens and neighborhood communities. Thus, citizen participation was becoming a central ingredient in Brazilian policymaking even before the Workers Party in Porto Alegre created its now globally diffused scheme for a participatory budget (Abers 2000; Figueiredo and Lamounier 1997). Accordingly, the 1981 legislation on the National System for the Environment included representatives from state and local-level environmental agencies on the board of newborn CONAMA. The participatory ethos of CONAMA, in turn, reshaped the organization of state and local-level environmental policy (Hochstetler and Keck 2007). CONAMAs openness to civil society creates an institutional path qualitatively different from that of other CONAMAs in the region; Brazilian authorities can claim to have in CONAMA the worlds largest environmental parliament.11

The past does not determine the future; it rather situates entrepreneurial institutional agency. Chile had been a Latin American leader during earlier waves of green state making. Its first Forestry Code dates from 1872 (the second one from 1931); first Forest Service, from 1873; and first protected area, from 1907the Malleco Forest Reserve (Camus 2006). But in the 1970s and 1980s, under Pinochet, neoliberal ideas became institutionalized, constraining the action of green-minded national elites. Someone like

11 http://www.mma.gov.br/conama/

Web End =http://www.mma.gov.br/conama/ . Accessed in January 11, 2011.

258 St Comp Int Dev (2014) 49:242265

Federico Albertthe institutional entrepreneur of state conservationism in late-nineteenth century Chilecould not have existed or become influential in those decades, which coincided with the early years of post-Stockholm green state diffusion. The case of Chile helps us understand the complexity of path-dependence. Chile had lead its Latin American neighbors during earlier waves of green state diffusion, just as it led its neighbors when it came to railway building, welfare, industrialization policies, and so on (Thorp 1998). But that did not mean that Chile had to become a leader in late twentieth century green state building. The Chilean CONAMA was not established until 1994. We might have expected Chiles tradition of positivist, centralized, and authoritarian government to translate into a strong green state, and this did eventually occur. But historical contingency also played a role, delaying the creation of the Chilean green state by several decades.

In the positivist and centralized Chilean state, policymaking is strongly influenced by what the policy community in universities, think-tanks, and research centers advocates. This relationship between state and society has well-established roots (Gngora 1986; Muoz 1993; Silva 2008). Thus, Chiles institutional entrepreneurs were not ultra-agentic technocrats, experts free of the influence of any social environment. Urban planners, civil engineers, and an assortment of scientists composed a solid epistemic community associated with the Center for Environmental Research and Planning (CIPMA). These professionals structured the arena for green state diffusion, producing the scientific discourse associated with environmental reform: CIPMA organized the Chilean Scientific Summit on the Environment in 1983, which took place every 3 years until 2002. In 1986, it started an information center, which was given to the newborn CONAMA 13 years later, now part of the library of the Ministry of Environment (Geisse 1993). That ministries and important state organizations commonly run libraries and documentation centers open to the public, as I found during fieldwork, is an unequivocal indicator of the tradition of state continuity and the public-mindedness of professional elites in Chile.

In elite-ruled Chile, the CONAMA could never have been as participatory as the original Brazilian model. The label was adopted, since it symbolized compliance with an international paradigm, but the regulatory system that unfolded was all Chilean (see Table 1 below). CONAMA was translated into a positivist bureaucratic system that invited only minimal participation from civil society. In fact, when the business community expressed fears about the abilities and political propensities of region-level bureaucrats, institutional entrepreneur Solari employed old institution of the intendente to sort things out. 12 The intendente is the natural agent (as the re-founding 1833 Constitution put it) of the President of the Republic at the local level. In the agreed legislation, the presidentially appointed intendente had veto power over the technocratic ruling of sectoral, region-level state regulators, who integrated region-level COREMAs. The meta-rule of the game, then, was that the President of the Republic, via the intendentes, could reverse decisions that had been made at the regional level if the President considered it appropriate. The translation of CONAMA into a system of de-concentrated bureaucratic autonomy and political centralization was no accident in light of state history. In addition to the symbolic legacies of modern state craft, there were real bureaucratic resources to build on. The region-level COREMAs

12 Interview with Solari.

St Comp Int Dev (2014) 49:242265 259

Table 1 The innovation and translation of green state blueprints: rise and fall of transsectoral regulatory/ coordinating agencies

Brazil Chile Peru

Domestic name for adopted global blueprint

SEMA CONAMA CONAM

Creation 1973 1990 1994

Original mandate Regulate/coordinate national environmental protection

Regulate/coordinate national environmental protection

Regulate/coordinate national environmental protection

End 1981 2010 2007

Merged into CONAMA Ministry of the Environment

Ministry of the Environment

Key state society features Federal, strong, bureaucratic state; participatory politics

Unitary, strong, bureaucratic state; elitist politics

Unitary, weak, patrimonial state; not so participatory, not so elitist politics

Final form CONAMA became the participatory council of environmental protection policy

CONAMA and region-level COREMAs became the system of technocratic environmental protection

CONAM became a toothless lobbyist of environmental protection

built on the region-level capacities of the state that had been strengthened during the dictatorship (Angell et al. 2001; Orihuela 2014) (Table 1).

The political economy of state making in Peru has different features. Perus political culture is much less positivist; instead, corruption and patrimonialism dominate modern state history (Cotler 1978; Basadre 1973; Quiroz 2008). No surprise then to find that Peru was always a laggard in environmentalist policy diffusion waves of the last 200 years. In 1962, under a modernization paradigm that favored the rational exploitation of resources, the National Office for the Assessment of Natural Resources (ONERN) attempted to carry out resource inventories, as the centralized development corporation CORFO in Chile and myriad state organizations in Brazil had been doing for at least two decades (Camus 2006; Thorp 1998). In parallel, a Forest and Hunting Service came out in 1963 (nine decades after that of Chile), along with the first Forestry Law (three decades after the second one of Chile and the first one of Brazil), which began increasing the number of protected areas. The first national park dated from 1957, the Tumbes National Forest, half-a-century after the Chilean equivalent (Dourojeanni 2009; ONERN 1974). The Peruvian state has had less control over its territoryits Amazon in particular, has produced much less state-sponsored developmentalism and, consequently, produced state conservationism much later as well.

Perus CONAM-led green state could not build on the capacities, bureaucratic culture, or legitimacy of a Ministry of the Interior or other equivalents. In particular, since there was little institutional legacy at the regional level to build on, and given that environmental regulation could delay investments urgently needed in an economy

260 St Comp Int Dev (2014) 49:242265

recovering from hyperinflation and political collapse, the policy choice was to centralize decision-making and leave the building of region-level state capacities for later. The assessment of state experts, even the environmentalists, was that region-level bureaucrats were not equipped to handle regulatory decisions, so a system centered in ministries at Lima was the only option from a pragmatic point of view.13 Despite multiple legislative and state organizational developments that put the formal Peruvian green state at regional benchmarks, bureaucrats had little autonomy and the national regulatory system remained largely toothless (Orihuela 2014).

Perus CONAM came to an end as it had come to life: suddenly and driven by soft-coercive and competitive mechanisms. In December 2007, President Alan Garca (20062011), who until then had shown no interest in making the environment a policy priority, announced the creation of a Ministry of the Environment, to guarantee to the world that Peru will fulfill its commitments.14 The sudden birth or death of state agencies and legislation characterizes modern state development in Peru. A good indicator of this is the extensive use of decrees of urgency, a legal form that gives the executive branch the capacity to legislate in the context of a national emergency. The document creating the Ministry of the Environment came along with close to 100 other decrees of urgency, nominally to ease the path to a free trade agreement with the USA, the exigent need of the time. In the last two decades, the number of such non-regular forms of legislation approved in Peru amounted to ten times those approved in Venezuela or Colombiawhile legally stable Chile does not use an equivalent device (Munck 2010). Such an institutional tradition does little to favor the continuity of bureaucratic schemes: ad hoc solutions come and go, and political actors employ extremely short-sighted timeframe. As a result, it is difficult for any Peruvian state organization to develop a rational bureaucratic culture.

Conclusion

To equate state institutional change with top-down diffusion is a reductionist and ultimately misleading metaphor. Instead, a bridging institutionalist framework makes us aware of the complexity and interplay of processes and mechanisms underpinning policy diffusion and state change. As shown, Latin American receptor sites did not just receive new ideas from abroad; they also produced or helped in the production of new institutional models, from the Brazilian Tijuca Forest, which pre-dated Yellowstone, to the Peruvian communal reserve, which predated the equivalent Brazilian scheme in the Amazon, passing by the local participatory politics imagery that shapes contemporary state making around the globe. Moreover, at Stockholm 1972, Rio 1992 and elsewhere, domestic actors (best understood as globalized nationalists), co-produced and legitimized global blueprints. So world society and global blueprints need to be unpacked in order to see how they are constructed and reconstructed, or else they risk becoming black-boxes for social analysis. In particular, domestic institutional entrepreneurs have agency. They are capable to translate and modify ideas received from abroad, and so nation-states are much more

13 Interviews with Mariano Castro and Pedro Solano.

14 La Repblica, December 21, 2007.

St Comp Int Dev (2014) 49:242265 261

than enactors of wider world cultural institutions (Frank et al. 2000, 99). Studying the actual transformation of an international blueprint into domestic practice, one finds that scripts are not played the same everywhere. Diffusion mechanisms, opportunity structures, and national institutional environments create an array of avenues for green state adoption and re-creation. After green state diffusion, Latin American states are still quite particular: converging divergence characterizes globalized institutional change.

Each institutions-matter framework has its explanatory virtues, yet by itself is insufficient. It is the interaction between the sociological and historical properties of institutions that rightly situates (not determines) rational policy agency. Even when several countries are moving in roughly the same direction, the institutions they create are likely to be different, as the case studies from Brazil, Peru, and Chile illustrated. All three states moved in roughly the same direction, but in markedly different ways. First, diffusion mechanisms lead us to expect divergence: when normative and mimetic mechanisms are relatively strong vis--vis coercive forces, reflecting the rising power of green-minded coalitions, formal state change is followed by more meaningful real state change. When the coercive mechanism rules unmatched, indicating that there is little ownership of reform from domestic elites, green state change ends up being formal for the most part. The latter is the case of green-laggard Peru producing a transsectoral regulatory agency in 1994.

Secondly, institutional entrepreneurs face shifting opportunity structures for political change; because these opportunities are never uniform, national experiences will differ. Opportunity, commonly brought by external events, but always in interplay with domestic processes, sometimes enhances, others constrains reformist agency, illustrating how the normative standpoints of a community of experts on their own are not enough to produce and qualify change. External events matter and commonly work as catalysts that can enhance existing dynamics, producing different processes of diffusion as a result. In Chile, FTA coercion was the ultimate catalyst of early-1990s reform, but green-minded social and political actors had penetrated power circles by then. Political economic coercion also catalyzed first a CONAM and later a ministry of the environment in Peru, but the diffused global blueprints met with a different domestic process, conditioning the production of a different, weaker green state variety. Again, homogenization only happened at the formal level.

Thirdly, national institutional environments provided contrasting domestic resources and cultures for the building of green states. Rooted institutional legacies interacted with politics of a more contingent nature, setting up a range of evolving contexts for policy diffusion. In Brazil, the structures of the modern civil service and state developmentalism met with the progressive opening of the military and the rise of participatory democracy. In Chile, the strong state, composed of various forms of centralized bureaucratic autonomy, met with two decades of Pinochetian neoliberalism. In Peru, weak state legacies met with the economic and political collapse of the late-1980s followed by the hegemony of a radical neoliberal coalition in the 1990s, constituting the most adverse context for green reform. Thus, innovation and translation are to be understood in social context. Still, despite structural and contingent conditions, neither the leader country nor the laggard ones had to follow their ultimate green state trajectories. Situated in particular institutional environments, facing critical events and everyday contingencies, relatively autonomous actors bent rules and molded arenas for policy diffusion and institutional change.

262 St Comp Int Dev (2014) 49:242265

Appendix

Table 2 Year of adoption of green state blueprints

National environmental law

Argentina 1973 1996 2002

Bolivia 1991 1993 1992

Brazil 1973 1992 1981

Chile 1990 2010 1994

Colombia 1993 1993

Costa Rica 1987 1995 1995

Cuba 1981 1994 1997

Ecuador 1996 1996

El Salvador 1997 1998

Guatemala 1986 2000 1986

Haiti 1994 1994

Honduras 1990 1996 1993

Mexico 1972 1994 1982

Nicaragua 1979 1994 1996

Panama 1990 1998 1998

Paraguay 2000 2000

Per 1994 2008 2005

RepublicaDominicana

2000 2010 2000

Uruguay 1990 1994

Venezuela 1977 1976

Transsectoral/ coordinating regulatory agency

Ministry of the environment

Acua 1999; Braes 1991, ministry websites and national legislation

References

Abers R. Inventing local democracy: grassroots politics in Brazil. Boulder: Lynne Rienner; 2000.

Acua G. Marcos regulatorios e institucionales ambientales de Amrica Latina y el Caribe en el contexto del proceso de reformas macroeconmicas: 19801990. Serie medio ambiente y desarrollo 20. Santiago: CEPAL; 1999.

Andrews R. Managing the environment, managing ourselves: a history of American environmental policy.New Haven: Yale University Press; 2006. p. 22832.

Angell A, Lowden P, Thorp R. Decentralizing development: the political economy of institutional change inColombia and Chile. New York: Oxford University Press; 2001.

Ban C. Brazils liberal neo-developmentalism: new paradigm or edited orthodoxy? Rev Int Polit Econ.2013;20(2):298331.

Basadre J. El Azar en la Historia y sus Lmites. Lima: Ediciones P.L.V.; 1973.

Bockman J, Eyal G. Eastern europe as a laboratory for economic knowledge: the transnational roots of neoliberalism. Am J Sociol. 2002;108(2):31052.

Braes R. Institutional and legal aspects of the environment in Latin America, including the participation of nongovernmental organizations in environmental management. Washington: Inter-American Development Bank; 1991.

St Comp Int Dev (2014) 49:242265 263

Callon M, Latour B. Unscrewing the big Leviathan: how actors macro-structure reality and how sociologists help them to do so. In: Knorr-Cetina, Cicourel. Advances in social theory and methodology: Toward an integration of micro-and macro-sociologies; 1981. p. 277303.

Campbell JL. Institutional change and globalization. New Jersey: Princeton University Press; 2004. Campbell JL, Pedersen OK. The rise of neoliberalism and institutional analysis. New Jersey: Princeton

University Press; 2001.

Camus P. Ambiente, bosques y gestin forestal en Chile, 15412005. Santiago: LOM Ediciones; 2006. Carruthers D. Environmental politics in Chile: legacies of dictatorship and democracy. Third World Q.

2001;22(3):34358.

Chorev N. Changing global norms through reactive diffusion: the case of intellectual property protection ofAIDS drugs. Am Sociol Rev. 2012;77(5):83153.

Conca K, Dabelko G. Green planet blues four decades of global environmental politics. Colorado: WestviewPress; 2010.

Cotler J. Clases, estado y nacin en el Per. Lima: IEP; 1978.

Di Maggio P, Powell W. The iron cage revisited: institutional isomorphism and collective rationality in organizational fields. Am Sociol Rev. 1983;48:14760.

Dobbin F, Simmons B, Garret G. The global diffusion of public policies: social construction, coercion, competition, or learning? Annu Rev Sociol. 2007;32:44972.

Dourojeanni M. Crnica forestal del Per. Lima: UNALM/UNMSM; 2009.

Drummond J. The garden in the machine: an environmental history of Brazils Tijuca Forest. Environ Hist1996;1(1):83104.

Drummond J, Barros-Platiau AF. Brazilian environmental laws and policies, 19342002: a critical overview.Law Policy. 2006;28(1):83108.

Drummond JA, Franco JLA, Ninis AB. Brazilian federal conservation units: a historical overview of their creation and of their current status. Environment and History. 2009;15(4):46391.

Dryzek J, Downes D, Hunold C, Schlosberg D, Hernes HK. Green states and social movements: environmentalism in the United States, United Kingdom, Germany, and Norway. Oxford: Oxford University Press; 2003.

Evans P. Embedded autonomy: states and industrial transformation. Princeton: Princeton University Press; 1995. Figueiredo R, Lamounier B. As cidades que do certo. Expericias inovadoras na administrao pblica brasileira. Brasilia: M.H. Comunicao; 1997.

Frank D, Hironaka A, Schofer E. The nation-state and the natural environment over the twentieth century. AmSociol Rev. 2000;65(1):96116.

Geisse G. 10 Aos de debate ambiental. Santiago: CIPMA; 1993.

Giddens A. Central problems in social theory: action, structure, and contradiction in social analysis. Berkeley:University of California Press; 1979.

Gilardi F. Transnational diffusion: norms, ideas, and policies. In: Carlsnaes W, Risse T, Simmons B, editors.Handbook of international relations. Thousand Oaks: SAGE Publications; 2012. p. 45377.

Gngora M. Ensayo Histrico sobre la Nocin de Estado en Chile en los Siglos XIX y XX. Santiago: EditorialUniversitaria; 1986.

Greif A. Institutions and the path to the modern economy: lessons from medieval trade. New York: CambridgeUniversity Press; 2006.

Guimares R. The ecopolitics of development in the Third World: politics and environment in Brazil. Boulder:Rienner Publishers; 1991.

Haas PM. Epistemic communities and international policy coordination. Int Organ. 1992;46(1):135.

Hall P. Historical Institucionalism in Rationalist and Sociological Perspective. In: Mahoney J and Thelen K, editors. Explaining institutional change: ambiguity, agency, and power; 2010; pp. 204223Hall PA, Soskice DW. Varieties of capitalism: the institutional foundations of comparative advantage, vol. 8.

Oxford: Oxford University Press; 2001.

Hall P, Taylor R. Political science and the three new institutionalisms. Polit Stud. 1996;44(5):93657. Hironaka A, Schofer E. Decoupling in the environmental arena: the case of environmental impact assessments.

In: Hoffman AJ; Ventresca MJ. Organizations, policy, and the natural environment: institutional and strategic perspectives; 2002. p. 214231.

Hirschman A. Journeys toward progress. Studies of economic policy-making in Latin America. New York:Twentieth Century Fund; 1963.

Hochstetler K, Keck M. Greening Brazil: environmental activism in state and society. Duke University Press; 2007. Hoff K, Stiglitz J. Modern economic theory and development. In: Meier G, Stiglitz J, editors. Frontiers of development economics: the future in perspective. New York: Oxford University Press; 2001.

Katz R, Del Favero G. Medio Ambiente en Desarrollo. Santiago: CEP; 1993.

264 St Comp Int Dev (2014) 49:242265

Keck M, Sikkink K. Activists beyond borders: advocacy networks in international politics. Ithaca: CornellUniversity Press; 1998.

Klyza C, Sousa D. American environmental policy, 19902006: beyond gridlock. Cambridge: MIT Press; 2008. Kohli A. State-directed development: political power and industrialization in the global periphery. New York:

Cambridge University Press; 2004.

Lacunza S. From dictatorship to democracy: environmental reform in Chile. Hast Int Comp Law Rev.1996;19:53995.

Latour B. Science in action: how to follow scientists and engineers through society. London: Harvard university press; 1987.

Li T. The will to improve: governmentality, development, and the practice of politics. Durham: DukeUniversity Press; 2007.

Mahoney J. Path dependence in historical sociology. Theory Soc. 2000;29(4):50748.

Mahoney J, Thelen K. A theory of gradual institutional change. In: Mahoney and Thelen, editors. Explaining institutional change: ambiguity, agency, and power; 2010. p.137.

McAdam D, Tarrow S, Tilly C. Dynamics of contention. New York: Cambridge University Press; 2001. McAllister L. Making law matter: environmental protection and legal institutions in Brazil. Stanford: Stanford

Law Books; 2008.

McCormick J. Reclaiming paradise: the global environmental movement. Bloomington: Indiana UniversityPress; 1989.

Meadows DH, Meadows DL, Randers J. The limits to growth: a report for the Club of Romes project on the predicament of mankind. New York: Universe Books; 1972.

Meyer JW, Rowan B. Institutionalized organizations: formal structure as myth and ceremony. Am J Sociol.1977;83:34063.

Meyer J, Frank D, Horonaka D, Schofer E, Tuma N. The structuring of a world environmental regime, 18701990.Int Organ. 1997;5:62351.

Migdal J. State in society: studying how states and societies transform and constitute one another. New York:Cambridge University Press; 2001.

Miller S. An environmental history of Latin America. New York: Cambridge University Press; 2007. Munck G. Repensando la cuestin democrtica: la regin andina en el nuevo siglo. Revista de Ciencia

Poltica. 2010;30(1):14961.

Muoz O. Polticas pblicas. Historias personales. Santiago: CIEPLAN; 1993.

Ocampo JA, Ros J. Shifting Paradigms in Latin Americas Economic Development. In: Ocampo and Ros, editors. The Oxford handbook of Latin American economics. OUP Oxford. NY: Oxford University Press; 2011. p. 325OECD. Environmental performance reviews: Chile. Paris: OECD Publications; 2005.

ONERN. Lineamientos de poltica de conservacin de los recursos naturales renovables del Per. Lima:ONERN/OEA; 1974.

Orihuela JC. The making of protected areas in the Peruvian Amazon. LASA Annual Meeting Paper; 2013. Orihuela JC. The environmental rules of economic development: governing air pollution in Chuquicamata and

La Oroya. J Lat Am Stud. 2014;46(1).

Pierson P. Politics in time: history, institutions and social analysis. Princeton: Princeton University Press; 2004. Pierson P. Public policies as institutions. In: Shapiro I, Skowronek S, Galvin D, editors. Rethinking political institutions: the art of the state. New York: NYU Press; 2006.

Polanyi K. The great transformation. New York: Rinehart; 1944.

Pulgar-Vidal M. Para Hacer Tortillas Hay Que Romper Huevos: Historia de La Gestin Ambiental en el Per:19902005. Lima: SPDA; 2006.

Quiroz A. Corrupt circles: a history of unbound graft in Peru. Washington: Johns Hopkins University Press; 2008. Rootes C. Environmental movements: local, national and global. London: Frank Cass Publishers; 1999. Schelling T. Micromotives and macrobehavior. New York: Norton; 1978.

Scott R. Approaching adulthood: the maturing of institutional theory. Theory Soc. 2008;37(5):42742. Sikkink K. Ideas and institutions: developmentalism in Brazil and Argentina. Ithaca: Cornell University Press;

1991.

Silva E. Democracy, market economics and environmental policy in Chile. J Interam Stud World Aff.1996;38(4):133.

Silva P. In the name of reason: technocrats and politics in Chile. University Park: Pennsylvania StateUniversity Press; 2008.

Strang D, Meyer J. Institutional conditions for diffusion. Theory Soc. 1993;22:487511.

Streeck W, Thelen K. Beyond continuity. Institutional change in advanced political economies. New York:Oxford University Press; 2005.

St Comp Int Dev (2014) 49:242265 265

Tarrow S. Power in movement: social movements and contentious politics. New York: Cambridge UniversityPress; 1998.

Tarrow S. The new transnational activism. New York: Cambridge University Press; 2005.

Taylor M. Institutional development through policy-making. A case study of the Brazilian Central Bank.World Polit. 2009;61(3):487515.

Thelen K. How institutions evolve: insights from comparative-historical analysis. In: Mahoney J, Rueschemeyer D, editors. Comparative historical analysis in the social sciences. New York: Cambridge University Press; 2003.

Thorp R. Progress, poverty and exclusion: an economic history of Latin America in the 20th century.Washington: IADB; 1998.

Tilly C. To explain political processes. Am J Sociol. 1995;100(6):1594610.

Van Cott D. Regional environmental law in the Americas: assessing the contractual environment. Univ MiamiInteram Law Rev. 1995;26(3):489533.

World Commission on Environment and Development. Our common future. New York: Oxford UniversityPress; 1987.

Jos Carlos Orihuela is Associate Professor at the Department of Economics, PUCP, where he teaches political economy coursework. He has wide research interests on the political economy of resource-based development and environmental change. His work has been published in the Journal of Latin American Studies, World Development, Journal of Environmental Studies and Sciences, and the European Journal of Development Research.