The contingency of Senate consent: a study of the determinants of roll call confirmation voting on executive branch appointments, 1945–1996

Abstract

I develop a model to explain U.S. Senate roll call confirmation vote tallys (i.e., the level of affirmative voting) for executive branch appointments submitted by the president from 1945 to 1996. The theoretical framework assumes senators as strategic actors operating within the confirmation context where many senators are cross-pressured between a norm of deference towards the appointing president which pervades the environment and their own policy/constituency agenda. The dampening effect on negative voting by this presidential leverage on appointments is attenuated when senators assess the situation and determine it is politically safer to vote to reject. The findings show that the grounds of opposition voiced against a nomination and aspects of the appointing president’s political strength are important components in explaining negative voting in confirmation roll calls.

Acknowledgements

The author wishes to thank the reviewers and editor of SSJ for their helpful comments. In addition, gratitude is due to Larry Berman, James Spriggs II, John Gates, Benjamin Highton, and Matthew Hoddie for their useful and insightful suggestions. An earlier version of this paper was presented at the 2000 meeting of the Western Political Science Association.

Notes

1 Technically, roll calls in the Senate are made necessary when 1/5 of those present request such a recording of a floor vote. In reality though, due to the Senate’s highly individualized orientation where each senator has so much power, if just one senator wishes a recorded roll call, other senators will respect her wishes and allow that 1/5 threshold to be reached. This practice is readily observed on nominations. There are no recorded roll calls on this motion for a recorded roll call. This question of why and under what circumstances a roll call on a nomination occurs is an interesting one but is beyond the scope of this study—and since they can be and many times are just a function of one or a very few senators desiring it, finding strong or telling regularities in that regard may be very difficult.

2 See Scigliano (1971), Songer (1979), Segal (1987), Felice and Weisberg (1988), Massaro (1990), Cameron et al. (1990), Abraham (1985), Segal et al. (1992), Overby et al. (1992), Ruckman (1993), Maltese (1995), Watson and Stookey (1995).

3 See Neustadt (1990), Schlesinger (1973), and Berman (1987).

4 Senate debates in the Congressional Record indicate that this is a common view articulated by senators, i.e., allowing the president general freedom to put into office his choices even when those nominees are ideologically objectionable in some manner.

5 See River and Rose (1985), Edwards (1989), and Collier and Sullivan (1995).

6 Cabinet offices consist of the executive department heads as well as other high-level posts that have been given cabinet-level status in the modern era—director of the Central Intelligence Agency, director of the Office of Management and Budget, chair of the Council of Economic Advisers, U.S. ambassador to the United Nations, director of the National Drug Control Policy, and U.S. Trade Representative. Lower executive posts are comprised of executive branch nominations that are beneath cabinet-level and are targeted to executive departments or to independent commissions, corporations, boards, or agencies.

7 Out of the total of 348 confirmation roll call votes in the period under consideration here, 111 were to the Cabinet, 237 were to the lower executive branch (122 of these were to executive departments and 115 were to independent agencies or commissions), and 65 were to the State Department (including both cabinet and non-Cabinet levels).

8 These three categories are collapsed into a single variable because they all are distinct public statements (with the accompanying calculations) by senators of support for and opposition against a president’s choice. Thus the dataset here consists of 318 separate nominations with 348 resulting confirmation votes—this is due to some nominations having several confirmation votes rendered upon them. For example, one can observe on one appointment an attempt to end a filibuster via invoking cloture (a filibuster constituting an overt attempt to kill a nomination—the cloture vote on Abe Fortas’ elevation to Chief Justice on the Supreme Court is the most prominent example), an attempt to recommit it back to committee (also an overt attempt to kill a nomination), and then the ultimate vote on confirmation. Multiple confirmation votes on one nomination are relatively rare.

9 One-tailed test of significance was utilized. Standard tests for multi-collinearity and heteroscedasticity were also used with no discernible problems in those results.

10 As just noted in the text, none of the 51 Year variables reached minimal levels of statistical significance and thus are not placed in the OLS results table due to space limitations.