ABSTRACT
This chapter explores challenges with causal inference for “long-term effects”. These are effects happening a long time after the intervention that contributed to them has ended. When conceptualizing this book, its editors and authors felt that it would be worthwhile to systematically think through how existing definitions and methods for causal inference can be applied to studying long-term effects, and what additional challenges exist compared to application over shorter time scales. This chapter makes this attempt. It explains what is meant by cause and effect, discusses two standard definitions of causality, and introduces methods for causal inference. It then investigates and makes conclusions about challenges associated with methods for causal inference, highlighting challenges associated with the long-term. Two methods are used to render this discussion concrete: a Quasi-Experimental Design using Propensity Score Matching, and Contribution Analysis.
