The study of presidential decision making strikes an uneasy balance between the parsimony of a particular presidential decision and the complexity of the organizational dynamics and political environment that shape that decision. Given this fundamental tension, and the potential scope of any model designed to fully represent the process, arriving at a generalizable understanding of presidential decision making is an especially vexing task. This undertaking is further complicated by the fact that presidential decision making varies substantially within and across administrations (Hult 1993). Simply determining what one means by presidential decision making and what unit of analysis to employ are difficult issues in themselves.
Given the myriad components of presidential decision making, how does one begin to generalize about the wide variety of decision-making systems and resultant outcomes visible across the presidential landscape? This article has a particular synthesis in mind. It will connect theory with two empirically rich case studies on economic policy making in the Eisenhower and Carter administrations to develop complementary notions of conditional partisanship and institutional responsibility. These concepts substantially account for the impact of strategic incentive-driven behavior and institutional constraints on presidential decision making. In doing so, the article considers the difficulties inherent in using rational choice as a link between presidential studies and political economy. In particular, the article demonstrates that rational choice assumptions about presidential behavior do not adequately explain empirical decision-making outcomes in political economy. The case studies provide a test of these assumptions and illustrate the need for a focus on the ways in which institutions constrain presidential response to exogenous incentives.
A fundamental cleavage (or tension) has emerged in the literature on presidential decision making in the past decade or so, one between decision making viewed as a function of presidential style and the same behavior posited as an interaction between political incentives and institutional structures. Shull (1999) provides a useful example of this tension; competing theoretical perspectives coexist uneasily in an end-of-the-century assessment of presidential decision making, as understood by almost two dozen leading presidency scholars. This article underscores the immense difficulty that scholars confront in attempting to move toward a more synthetic notion of presidential behavior.
Regarding the presidential-style approach, Moe (1993) notes that traditional notions of the presidency pass the institutional structures of decision making through a presidential filter (p. 346). The system is viewed as highly malleable, with its structures continually redefined by subsequent presidents, as a function of an individual president's personality, values, beliefs, background, and style; the president essentially fashions the structures of the institution in his own image. The result is a highly particularized (or perhaps balkanized) understanding of presidential behavior across administrations, one that is largely resistant to theoretical generalization. To the extent that the presidential-style approach is generalizable, it usually suggests that a president selects his decision-making apparatus from among a number of recognized types--formal, competitive, collegial, and so forth--as best meets his needs. Walcott and Hult (1995) have referred to this theoretical perspective (in the specific context of presidential advising) as a personal-contingency approach.
A number of scholarly works, written in the 1970s and 1980s, adopt this personal-contingency (or management-style) approach To understanding presidential decision making. For example, George (1980) employs a cognitive psychological approach to presidential behavior, defining it in part as a function of "character-rooted needs" and "psycho-dynamic patterns for adapting to stressful experiences" (p. …