Policymaking in the American States, 1787-2020

Abstract

This paper combines tools from natural language processing and structural estimation to analyze the adoption and diffusion of policies across U.S. states over the period 1787-2020. We employ a large language model to extract policy content from all state statutes enacted during this period (N=2.5 million). Using an unsupervised clustering algorithm applied to vector representations of policy content, we group related policies enacted across states with the goal of measuring state legislative activity on these policy clusters. After validating this measurement procedure, we estimate a discrete-choice model of policymaking to analyze the determinants of policy diffusion across states. States with greater geographic, economic, and political similarity implement more similar policies. While polarization has risen significantly since the 1990s, consistent with prior literature, we show that current levels of polarization are comparable to those observed in the pre-WWII period, following a U-shaped pattern over time. This rise in polarization has primarily come at the expense of similarity across geographic lines. Polarization primarily reflects differences in policy choices within topics, rather than from states selecting different topics to address. Finally, we document an increasing nationalization of policy, with federal legislative texts exerting growing influence on state legislation since the post-war period.