Causal Inference from Pivotal Politics Theories

Abstract

We review attempts to evaluate Pivotal Politics and related theories in light of the social sciences’ movement toward causal empiricism. To accommodate limitations of available data, most existing empirical approaches require additional assumptions about key variables. Moreover, it is difficult to find sources of exogenous variation in causal factors of interest. Although causal empiricism and pivotal politics theories share a focus on counterfactual comparisons, we argue that many design-based strategies for inference are ill suited to directly testing the theory’s predictions. We conclude by discussing opportunities to apply the contemporary causal inference “toolkit” to evaluate the theories’ central predictions, highlighting both the barriers to application as well as potential avenues for future use.