The last three decades of environmental policymaking have generated some very ambitious and comprehensively formulated policies, ranging from the local to the global level. These policies often have the explicit aim of addressing and reversing some of the most significant trends of environmental degradation, including biodiversity loss and climate change. However, despite evidence of successful progress in the formulation of environmental policies, their implementation and goal fulfillment are generally low. The inadequate fulfillment of environmental policy goals has partially been linked to the perceived limitations of using coercive-based implementation to solve complex environmental problems. Meanwhile, participation and learning are gaining prominence, since they are expected to cope better with complex and unpredictable environmental systems and thereby contribute to the implementation of environmental policy. However, complexity is only one of several decisive contextual factors, which affect environmental policy implementation and governance. For this reason, this thesis questioned and explored the extent to which learning constitutes the most appropriate mode for environmental policy implementation, and under what circumstances.
The introductory chapter frames learning as an implementation mode and conceptualizes it as one of three ideal-type implementation modes; the others being coercion and market. It introduces conflict as a crucial context for environmental policy implementation and hypothesizes that there are three facets of conflict, which are particularly relevant: conflicts of interest between stakeholders, conflicts in environmental policy goals, and legislation. Furthermore, it combines identified variables and concepts into a conceptual model that treats change in target group behavior as the outcome of policy and a dependent variable in environmental policy implementation. Policy outcomes are assessed in terms of their effectiveness and coherence.
The thesis was guided by the following research questions:
1. How do stakeholder conflicts of interest and legislation influence the effectiveness of learning as an implementation mode?
2. How are conflicts in environmental policy goals manifested, and what role can a learning-based implementation mode play in increasing policy coherence?
A multi-case study approach was chosen for this thesis since it allows for a detailed examination of the relevant contextual factors and an exploration of the possible causal mechanisms. As the thesis hypothesizes that conflict is a crucial context for environmental policy implementation, the case studies were selected in two environmental policy domains with different levels of conflict. Chapters 2-4 comprise several cases of environmental policy implementation that involve high conflict, concerning river restoration that affects hydropower production. These cases span different governance scales in Sweden, the United States and the European Union. Chapter 5 details another case, containing low conflict, concerning learning as an implementation mode in adaptations of urban water services related to climate change in the Stockholm Region, Sweden.
The thesis uses a variety of data collection strategies; including document analysis, semi-structured interviews, and participant observation. This allows for a strong contextual understanding and for triangulations of the insights gathered from the various data sources. The main paths for establishing the claims of causation, between policy initiatives, implementation modes and policy outcomes in the examined cases, are the counterfactual as well as the mechanism and capacities approaches.
A number of key findings emerge from the preceding chapters, related to the two thesis research questions. As concerns the first question; the thesis finds that legislation is a key determining element that influences the effectiveness of learning in situations that exhibit high levels of stakeholder conflicts of interest. Chapter 4, a case study of a failed learning-based intervention in the Ljusnan River basin, Sweden, demonstrates how the effectiveness of learning was limited by existing legislation since it gave the target group of policy the option to reach its objectives unilaterally. In contrast to Chapter 4, Chapter 5 shows how learning can produce effective policy outcomes, despite limited legislation, in situations of low stakeholder conflicts of interest. In this case, despite the absence of rules and regulations requiring climate change adaptation, the urban water service organizations in the Stockholm Region have enacted significant behavioral changes geared towards climate change adaptation through organizational learning.
As concerns the second research question; the results illustrate how environmental policy goal conflict is mainly materialized and manifested during the implementation of environmental policy. Chapter 2 concludes that, despite limited evidence of conflictual interactions at the level of policy goals and instruments, potentially strong conflicts emerge when it comes to policy implementation at both EU and member state level. The results also point towards the potential pitfalls and possibilities of the role that learning can play in policy coherence, largely depending upon legislation. Chapter 3 illustrates the differences, between Sweden and the United States, in the legislative settings of hydropower production as it concerns river restoration. Whereas hydropower production in Sweden is regulated by perpetual, property-like permits; in the United States, non-federal facilities are regulated by temporal licenses granting a time-bound privilege to use public lands and waters. Chapter 4 concludes that the perpetual hydropower permits in Sweden were an important reason behind the failed learning-based intervention in the Ljusnan River basin. The same chapter, however, also introduces the example of the Penobscot Basin, in the United States, where a learning-based intervention has produced significant river restoration outcomes while preserving hydropower generation at previous levels. Chapter 4 identifies the legal arrangements surrounding non-federal hydropower facilities in the United States, notably time-bound permits, as an important reason as to why the learning-based intervention in the Penobscot Basin produced results that increased policy coherence between different environmental policy goals.
Based on these answers to the two research questions, an overarching insight from the thesis is that the shape and formulation of legislation constitutes a significant variable, in determining the appropriateness of learning in environmental policy implementation that contains high conflict. The thesis’ analysis suggests that learning, together with favorable legislation, could be a viable way of dealing with complexity, while preventing the inaction that may arise from stakeholder conflicts of interest.
Through its empirical cases, this thesis provides evidence and arguments that relate observed behavioral changes to the policy of interest, in varying degrees. Moreover, the examples, in the discussion of environmental policy implementation affecting fisheries and agriculture, show that the findings are also relevant to situations of environmental policy implementation that affect the industrial use and extraction of natural resources more broadly. The thesis ends by offering a suggestion to policymakers who are faced with ineffective implementations of environmental policy: that they recognize the requirements and limits of learning, particularly in high conflict situations.