Nathan D. Gibson
The Politics of Presidential Centralization
Presidential centralization of policy creation is a surprisingly difficult to measure and understudied presidential tool, especially considering its foundational role in presidential management of the executive branch. There are no existing measures of centralization by policy area, leaving at least two major gaps in the literature: namely, how centralization varies across policy areas or how centralization and politicization relate. This paper draws empirical predictions from a formal theory that jointly models centralization and politicization and tests them using the Survey on the Future of Government Service, a large-scale survey of federal government executives. With these surveys, I introduce the first two measures of centralization by policy area to examine how presidents strategically engage in centralization and politicization. Among other results, I show that greater ideological distance between the president and an agency's careerists is associated with increased centralization. Furthermore, politicization is replaced by centralization as ideological distance grows.
The Strategic Contexts of Presidential Centralization
Presidents consider a broad array of factors when deciding when to centralize policy creation into the Executive Office of the President. This paper examines how the broader, time-variant political context affects when presidents choose to engage in centralization, based on archival research of 352 policies randomly selected from the Eisenhower through Clinton Administrations. I use these data to test predictions about centralization inspired by an original formal model of the president's centralization-politicization decision. First, I find that Congress clearly impacts the centralization decision, as presidents are more likely to engage in centralization under conditions of divided government or greater ideological disagreement with either chamber of Congress. Within the executive branch, greater centralized staff capacity is distinct from, but associated with, higher levels of centralization. I also find evidence that Republican presidents are more likely than Democratic presidents to centralize governmental reorganization efforts.