When Is Hindsight 20/20? Public Officials’ Acknowledgement of Policy Failure

• In Event: Communication and the Media in State Politics

Abstract

One justification for federalism is that states can operate as policy laboratories, experimenting to innovate the most effective policies (Brandeis 1932). We have an extensive literature on when new are adopted (e.g., Berry & Berry, 1990) and how they diffuse across state lines (e.g. Dobbin, Simmons, & Garrett, 2007). However, this scholarship assumes that experimenters always produce successful policies and has yet to explore the reactions to data suggesting a policy experiment has failed to produce the desired results. This paper examines the conditions under which state-level public officials acknowledge policy failure, which I define as a mismatch between the documented outcomes of a policy and the original legislation’s stated intent. Using a comparative case study of tax policy and juvenile justice policy in Texas and Washington, I make two main arguments: first, that a state’s bureaucratic capacity to collect data is distinct from its capacity to analyze it; consequently, these two bureaucratic capacities can vary independently of one another. Second, that the different combinations of collection and analytical capacity influence the likelihood of failure acknowledgement. My findings speak to the impact of state bureaucratic design and activity on state-level evidence-based policymaking.

Author

• Sarah Elizabeth James, Harvard University