Political appointees and auditors of politics: Essays on oversight of the American bureaucracy

This dissertation focuses on strategic oversight of the American bureaucracy. The first essay examines whether political appointees who are called to testify more frequently in front of hostile congressional hearings have shorter tenures. Cox Proportional Hazard models on a data set of Senate-confirmed appointees who entered or left a position between 1981 and 1991 suggest that a higher rate of appropriations hearings increases appointee tenure and that a higher rate of oversight hearings decreases tenure. The essay also considers which institutional arrangements may promote longer tenure and how appointees might use government positions as stepping stones to better jobs in the future.; The second essay develops a two-period principal-agent model to analyze how auditors may build and use their reputations in selecting investigations of policy programs. At the start of the first period, the legislature hires an auditor, who may be partisan (favoring either the Democrats or the Republicans) or nonpartisan. The auditor learns, but the legislature does not, the actual amounts of waste for a Republican and Democratic project. The auditor chooses to investigate one of the projects and reports the waste level to the legislature. At the start of the second period, the legislature decides whether to keep the original auditor or obtain a new one. The auditor then chooses between another Republican and Democratic project and reports the waste level. I extend the analysis by including a cost to firing the auditor and incorporating an election between the two periods to determine the legislature's partisan affiliation.; The third essay examines how members of Congress use the General Accounting Office to advance their own policy preferences and how the GAO chooses to investigate policy programs on its own. Using information from the GAO Documents Database for 1986–1997, I find that House committee chairpersons are more likely to request a GAO investigation when there is divided government. I also find that after the Republicans gained control of Congress in 1994, the GAO performed almost no defense investigations on its own initiative, but did not substantially increase investigations of projects that could be perceived as Democratic.