Search

This article explores how policy learning can improve comparative policy analysis by focusing on causality in learning processes. After summarizing the comparative credentials of the policy learning literature, the article outlines a framework of four learning modes, relating it to three approaches of causality: deterministic, probabilistic, and set-theoretic. It then builds on this to explore different approaches to causation and learning in relation to: policy change, political contexts, and, finally, the temporal and spatial dimensions of comparative policy analysis. The article concludes by showing how these challenges are addressed and suggesting implications for further research. [ABSTRACT FROM AUTHOR]

It seems paradoxical to suggest that theories of learning might be used to explain policy failure. Yet the Brexit fiasco connects with recent approaches linking varieties of policy learning to policy pathologies. This article sets out to explain the UK government's (mis)management of the Brexit process from June 2016 to May 2019 from a policy learning perspective. Drawing on interviews with UK policy-makers and stakeholders, we ask how did the UK government seek to learn during the Brexit negotiations? We consider four modes of learning: reflexivity, epistemic, hierarchical, and bargaining. By empirically tracing the policy process and scope conditions for each of these, we argue that learning through the first three modes proved highly dysfunctional. This forced the government to rely on bargaining between competing factions, producing a highly political form of learning which stymied the development of a coherent Brexit strategy. We argue that the analysis of Brexit as a policy process – rather than a political event – reveals how policy dynamics play an important role in shaping the political context within which they are located. The article concludes that public policy analysis can, therefore, add significant value to our understanding of Brexit by endogenising accounts of macro political developments. [ABSTRACT FROM AUTHOR]