Essays on the Political Power of Bureaucrats

In the first essay, I move towards a new theory of presidential policymaking. Unilateral presidential directives often face implementation problems in the executive branch. I argue these directives can be more fruitfully studied as instances of delegation. I present a theory of delegation within the executive branch, modeling the conditions under which the president is likely to delegate---and provide discretion---to administrative subordinates outside the Executive Office. This theory demonstrates that members of Congress benefit from agency discretion when the President acts alone. I show that it is often optimal for the President to knowingly permit agencies to deviate from a directive's mandate---in some cases, delegating to agencies insulated from presidential control. Ultimately, the model demonstrates how the politics of direct action are influenced by the necessity for bureaucratic cooperation.

The second essay provides an empirical investigation of agency fulfillment of daily, informal requests from members of Congress. I ask a fundamental, often considered question: what makes agencies more (or less) responsive to elected principals? To investigate this question, I leverage a dataset of over 20,000 congressional requests made by individual members of Congress to 12 executive agencies between 2007-2015. I find that executive agencies systematically prioritize majority party legislators but that this effect can be counter-acted when presidents politicize agencies through appointments. An increase in politicization produces a favorable agency bias toward presidential co-partisans. This same politicization, however, has a net negative impact on agency responsiveness; agencies are less responsive to members of Congress, but even less responsive to legislators who are not presidential co-partisans. The results suggest that presidents have the capacity to influence the flow of information between Congress and the bureaucracy.

Finally, in the third essay, I investigate an additional avenue of responsiveness: the geographic distribution of billions of dollars in federal grants. Recent studies find evidence that presidential preferences influence the allocation of federal spending. Yet these findings leave two largely open questions: how those outcomes are achieved and what (if any) role Congress plays in a presidency-centered understanding of distributive politics. I investigate both questions through a detailed analysis of grant allocation by the National Science Foundation (NSF) and Department of Energy (DOE) from 2007-2014. The former case provides a ``hard test'' for presidential influence due to the political insulation of the NSF, whereas the latter presents a case that ought to be consistent with existing findings of ``presidential particularism.'' Both allow me to leverage novel data on grant-related correspondence between the agency and members of Congress to provide an initial test of the efficacy of congressional casework. Contrary to standard expectations related to bureaucratic structure, I find evidence of political allocation in the NSF. Contrary to standard expectations related to presidential particularism, I find no evidence of political allocation in the DOE.