Advocacy Coalition Framework Petr Ocelík ESSn4007/ MEBn4001 Outline • Advocacy Coalition Framework • Case studies Advocacy Coalition Framework Policy process • Policy process: a process through which the public policy (or its components) is produced, terminated, or revised • Policy process is shaped by: 1. interactions of diverse actors influenced by institutional structures (Ostrom 2014; Sabatier 1988) 2. policy discourses and frames (Shanahan et al. 2011) • (number of more general structures and events) • Different policy process theories tend to emphasize different dimensions of the policy process Agency: bounded rationality • Individuals are goal-oriented but have limited time, resources and cognitive ability to consider all information, solutions, etc. (Simon 1957, Cairney 2012) → they use heuristics to make “good enough” decisions • Individuals rely on beliefs to decide to which information pay attention → Individuals (actors) tend to act according their enduring beliefs rather than according their short-term rational interests Belief system • Actors related to the world through perceptual filters (heuristics) based on enduring beliefs • Assimilation bias: belief systems condition actors to accept and interpret policy-relevant information in way that supports their beliefs Policy process and advocacy coalitions • Policy process involves (1) diversity of actors and their groups and occurs (2) mostly at the level of a policy subsystem – subset of political system defined by issue area • Actors perceive policy problems through a system of policy beliefs and struggle to translate their beliefs into policies • Advocacy coalitions (1) share policy beliefs and (2) coordinate their efforts • Dominant vs. minor coalitions • Principal vs. auxiliary coalition members • Policy brokers Policy subsystem • Policy subsystem is a subset of political system defined by particular issue area (Weible et al. 2016). • # of coalitions, patterns of coalition’s beliefs and coordination → three different types of subsystems Policy change • Major PC: changes in the core aspects of the policies • Minor PC: changes in the secondary aspects of the policies Four pathways to policy change: 1. External events: changes in government, disasters, crisis, etc. 2. Internal events: actor collapses, corruption affairs, etc. 3. Policy-oriented learning: gradual change in coalition beliefs reflecting new information 4. Negotiated agreement: resulting from collaborative institutions or hurting stalemate Advocacy Coalition Framework: A Case of Czech Coal Policy ⚫ brown coal production accounts for 46% of TPES and 51% of electricity mix ⚫ it is concentrated in the Sokolov Basin and the North Bohemian Basin ⚫ the territorial mining limits has been established by government decree in 1991 stakes: ⚫ a lifting of “the limits” became a key issue in energy policy since then ⚫ transition pathway to decarbonized economy very much depends on the future of coal Case study: Czech coal policy • Defined by (1) competing coalitions with (2) low intercoalition belief compatibility and (3) high intra-coalition and (4) low inter-coalition coordination (Weible et al. 2010: 524) • Further expected: coalitions compete for access to decision-making • Further expected: (some) experts are principal allies or opponents of the coalitions → high political use of expert info by coalitions Adversarial subsystem (Weible et al. 2018) 1. Shared policy core beliefs • normative assumptions on how specific policy field ought to be organized • captured by 4 Likert-type scales: • economy: costs/benefits of coal, regional development • environment: environmental and health impacts • policy: future of coal in energy mix, question of the mining limits • process: trust among key actors, regulatory framework 2. Factions • cohesive parts of a network • groups of actors that are connected more among themselves than with others Advocacy coalition detection • organizational actors involved in coal policy subsystem • the survey instrument (a self-administered online questionnaire) collects data on attribute variables: (1) policy core beliefs and (2) network ties sector responded total response rate (%) central and regional governance 16 16 100 central and regional political parties 16 18 89 environmental non-governmental organizations 8 9 89 research organizations 14 16 88 professional associations & trade unions 3 7 43 industry 11 17 65 total 68 83 82 Data collection network tie political influence (PI) network directed binary tie expert information (EI) exchange network directed binary tie political cooperation (PC) network directed binary tie policy core beliefs: • coal as a basis of economic growth • should be part of future energy mix • mining limits should be rescinded • legislative framework and stakeholder engagement are adequate • Led by state-owned energy company and Ministry of Trade and Industry • Highly influential with direct access to decision-making consists of 16 organizations: • 2 state agencies (central) • 1 regional agency (Ústí region) • 2 political parties (central) • 3 political parties (Ústí region) • 1 research organization • 5 companies, 2 NGOs Usual suspects: Industry coalition consists of 17 organizations: • 8 ENGOs • 2 state agencies (central) • 1 political party (central) • 6 research organizations policy core beliefs: • coal mining has severe enviro impacts • should not be base for future energy mix • mining limits should not be rescinded • legislative framework and stakeholder engagement are not adequate • Consists mainly of ENGOs and research organizations • Emphasis on relational capacity as well as expert knowledge Usual suspects: Environmental coalition Polarized policy core beliefs distribution • Decision-making actors (DMAs): competent ministries and ruling (central and regional) political parties • Key DMAs – three competent ministries – belong to different groups Industry Coalition = blue Environmental Coalition = green residual group = grey node size = reputational power Fragmentation of the decision-makers ACF: Coal policy in Czechia Two competing coalitions in a fragmented political system (Ocelík et al. 2019) • Expert information is crucial for management of complex sociotechnical systems (Giddens 1990) – includes evidence-based policy-making • Its importance increases under conditions of uncertainty (Cairney et al. 2016) • Two opposing approaches: • Technocratic governance: exp info abrades ideological differences and “builds bridges” • Expertise politics: exp info is used to defend ideological positions of their holders/providers Use of expert information • Block model (BM) is a simplified representation of a network (White et al. 1976): • Groups of nodes with similar relations to others (blocks) • Patterns of relations among blocks (social roles) Block modeling • Coalitions identified based on political cooperation and shared policy core beliefs Blocked density matrix: expert information Adj R^2 = 0.102 Bolded cells indicate significant differences from the average (network density = 0.173) Expert information: Tell me, I am right? • expert information is crucial for management of complex sociotechnical systems (Giddens 1990) • evidence-based policy-making • its importance increases under conditions of uncertainty • technocratic governance: exp info abrades ideological differences and “builds bridges” • expertise politics: exp info is used to defend ideological positions of their holders/providers • more than 2.5 times more likely to exchange expert information within advocacy coalitions than between the coalitions ➢ contributes to polarization and limits policy change by learning Expert information: Tell me, I am right? • Two adversarial coalitions detected • Support for a fragmentation of the decision-making actors → limits formulation of coherent policies • Expert info exchange strongly overlaps with the coalition patterns → limits policy learning between coalitions • Altogether, findings support the thesis on contestation of the transition process • Expectation: major policy change rather due to external factors such as the EU’s regulation and macro-economic trends Main findings (Ocelík et al. 2019)