Control without Confirmation: The Politics of Vacancies in Presidential Appointments
My dissertation, a book project, examines the politics and policy implications of vacancies and interim appointments in the U.S. bureaucracy. Presidents face a common adversary as they work to advance their policy priorities: time. As legislation crawls through an increasingly intransigent Congress, administrative policymaking offers a pragmatic alternative. Just over 1,200 top appointments (those requiring Senate confirmation, known as PAS positions) direct the politics and policy-making apparatus of modern administrative presidencies. Given their impact and the time constraints of a four-year term, one might expect PAS positions to be sporadically vacant at worst. This, however, is not the case. More than eight months into the current administration, 80 percent of major PAS positions were vacant. And vacancies are by no means a new phenomenon: over the past five administrations, PAS positions were vacant between 15 and 25 percent of the time. Since existing theories have focused on nominees and confirmed appointees, we have yet to consider the value of their absence for presidents' appointment strategies. My research does exactly that. I combine an innovative theoretical model with new data to study a question that has been largely ignored in executive politics research: how do presidents use vacancies in their top appointments to pursue their political and policy agendas?
I begin to answer this question with a theory of the strategic value of vacancies to presidents. As theories of appointments focus on the choice of nominee and the Senate's role post-nomination, they pay little attention to the president's decision to submit a nomination in the first place. Rather, scholars have considered vacancies as the mechanical byproduct of changes in administration or senatorial delays in confirmation, with no focus on how strategic presidential behavior might be accounting for these patterns. Importantly, vacancies in this context are more appropriately identified as PAS positions without a confirmed appointee, which includes positions that are unfilled and those filled with interim appointees. This coarse definition fundamentally restricts our ability to draw conclusions about how interim appointees and unfilled posts vary across (and within) administrations and agencies. Moreover, since existing models of the appointments process do not incorporate this distinction, we have little sense of how they operate theoretically. Consequently, I introduce a more nuanced classification that differentiates between empty and filled -- with interim or confirmed appointees -- positions and establishes a reversion point set by the president that better represents the status quo in negotiations over PAS appointees. As a contribution to our understanding of appointments, my formal model offers one of the first attempts to theoretically integrate this distinction while also incorporating the Senate's leverage to veto a nomination and the president's power to choose not to submit one in the first place.
My theory offers several testable hypotheses. First, I hypothesize that a vacant position will stay empty only when the position is high-skilled (i.e., positions that require more expertise, have more room to influence policy outcomes, and advance a larger political agenda) and the president prioritizes policy contraction, no matter the priorities of the Senate. This implies that, if the president prefers to shrink the policy reach of an agency, persistently empty high-skilled positions will occur even in unified government. Second, the president's choice to submit a nominee hinges on the cost of bargaining, conditioned by the Senate's value for oversight and time to confirmation. Lastly, presidents appoint interims no matter the position skill level but do not fill empty posts immediately with interims when prioritizing contraction. This implies that as presidential policy agendas slant towards expansion, the number of interims appointed to vacant positions will increase.
To assess the empirical implications of my model, I built an original dataset of PAS appointments including vacancies, nominees, and interim and confirmed appointees (more than 32,000 observations) from 1976 through 2016. I gathered these data from archived government reports processed using optical character recognition (OCR) software, files obtained through FOIA requests to the Office of Personnel Management and the Government Accountability Office, and nomination records scraped from Congress.gov. I also collected data on position policy jurisdictions, the capacity of position to control policy, and policy priorities to construct the main variable of interest, position value, for all PAS positions during this period, as well as data on each of my model's parameters and controls. Together, these data sources represent the most comprehensive data to date on appointments to and vacancies in PAS positions. My empirical strategy examines the reality of vacancies in two parts: a correlational study, and a causal inference one using a quasi-experimental design. My correlational approach exploits the nested structure of the data (i.e., positions within agencies within administrations) with a multilevel choice model. From this analysis we learn the impact of my theory's key parameters -- position value and policy priorities -- on presidents' choices to leave positions empty or appoint an interim official. To determine whether the option to appoint temporary leadership triggers presidents' use of interim appointees, I employ a difference-in-differences research design that exploits institutional variation in the procedural regimes governing vacancies. From this analysis, we gain a better sense of the strategic value of vacancies and whether presidents use them as my theory predicts.
At its core, my dissertation aims to identify the politics of vacancies in PAS positions within the separation-of-powers context and begin to understand the implications of persistent vacancies for administrative presidencies. In particular, it posits that vacancies are calculated choices that presidents make as a part of their larger strategies when bargaining with the Senate over confirmation. To do so, it first investigates the conditions under which presidents maintain empty posts or unilaterally-appointed interims without formal nominations. Second, it investigates conditions under which the Senate leverages its veto of a nomination to curb the president's use of interims or empty posts. This is a starting point for incorporating the political reality that PAS positions regularly remain empty or filled with interim appointees into research on executive politics and separation of powers. Lastly, it posits that the strategic use of vacancies has identifiable implications for administrative policy-making like implementation activities and regulation. To do so, it investigates conditions under which presidential appointment strategies achieve their expected policy outcomes. Appointments to the bureaucracy are presidents' most important instrument of political control and policy formation. Our current understanding of the extent to which these appointees matter for policy outcomes is limited by our inattention to decisively empty posts or unilaterally appointed interim officials. By considering the strategic value and policy implications of vacancies, my dissertation project engages research on executive appointments and political control of the bureaucracy, contributes to the growing literatures on presidential unilateral action and legislative obstruction, and speaks to work on separation of powers more generally.