Do Vacancies Hurt Federal Agency Performance?

The combination of the high workload associated with keeping top executive branch positions filled and political dysfunction has led to longer and more frequent periods of vacancies in the U.S. executive branch. While scholars commonly claim that such vacancies are harmful for performance, this claim has been difficult to evaluate because of theoretical disagreement, conceptual confusion, and measurement challenges. In this paper we evaluate the relationship between vacancies and performance, describing primary mechanisms by which vacancies (as opposed to turnover) influence performance. We conduct a cross-sectional study using new data on appointee vacancies during the Trump Administration and original performance data from a 2020 survey of federal executives. The Survey on the Future of Government Service includes questions designed to measure comparative self-reported agency performance and questions targeting the mechanisms hypothesized to link vacancies and performance. The paper includes efforts to define and validate the measure of performance, assess the directionality of the relationship between vacancies and performance, control for potential confounders that may explain both vacancies and performance, and evaluate the mechanisms by which vacancies negatively affect performance. The results from OLS models suggest that persistent vacancies are correlated with lower performance. In particular, agencies with persistent vacancies (e.g., 3-4 years) have performance ratings of about one standard deviation lower than those agencies with consistent confirmed leadership. The most likely mechanisms leading to these results are the effect of vacancies on leader time horizons, agency morale, and investment by key stakeholders. We conclude with implications for appointment politics and administrative politicization