Four models of causation in the design of anti-corruption policies

Abstract

Policy problems and government needs do not always match, hence a persisting gap between scholars and practitioners in the plethoric literature on policy design. Yet when it comes to explain a policy outcome or to anticipate the effects of a policy agenda we find ourselves between a rock and a hard place: we cannot explain anything with any method, we have to chose a model of causation and stick to the right method. Within the policy design framework, scholars provide a great diversity of theories and proposals for practitioners thereof. How can these actually contribute to design better policies? Do they even talk to one another? To answer these questions, the present chapter draws on a definition of the policy design framework based on the fourfold model recently proposed by B. G. Peters (Policy Problems and Policy Design, E. Elgar Publishing, 2018), namely causation, instrumentation, evaluation and intervention. We focus on causation, and contend that four theories of causality are actually handled by scholars and practitioners, which are grounded in different ontologies and have different methodological implications for research and practice. We review these models of causation one by one, following a typology of methodologies by P. T. Jackson (The conduct of inquiry in international relations: Philosophy of science and its implications for the study of world politics. Routledge, 2016). We start with the dualist ones (neo-positivist and realist), which explain causality in terms of regularity and necessity, then we pursue with the monist ones (analyticist and reflexive), which interprete causality in terms of configurational and contextual situations. We utilize research on anti-corruption policies as an illustration of these models of causation. This chapter concludes with a discussion about the problem of ontological alignment for policy design.