Despite the absence of scientific stipport for a tripartite alliance in defense poticymaking, political scientists and the general public continue to subscribe to a subgovernment exptanation. This article addresses the subgovernment tUtemma as it appties to the area of weapons procurement.
More precisely, it asks whether the subgovernment model accuratety describes and adequatety exptains the decisions to devetop and buitd major weapons systems. In answering this question, the criticat assumptions or propositions that comprise the model are tested, relying upon pubtished research and original data dram from 19 cases of military hardware decisions. The findings do lend some qualified support for a military subgovernment.Students of policymaking continue to rely upon some variant of a subsystem model to explain much of policy outcomes (McCool, 1990). The popularity of the model is evident particularly in the defense policy area, considered a classic case of subsystem politics (Mayer, 1991). Yet there is little empirical evidence to support this conceptual framework (Hamm, 1983). What support there is for this popular model is generally descriptive, impressionistic, and anecdotal. Still, there is the sense that defense policymaking is, at its core, a reciprocal dynamic involving the military services in the Pentagon, defense contractors, and the congressional Armed Services committees, guided by personal and organizational benefits.This article examines the subsystem model as it applies to the area of weapons procurement. More precisely, it asks whether the model accurately describes and adequately explains the decisions to develop and build major we^x>ns systems. In an effort to answer the question, the study begins with a review of the literature from which the critical propositions that comprise the model are gleaned.' The article then develops indices for examining the framework. That section also contains a description and brief discussion of the formal roles and powers of each set of actors in the defense subsystem dynamic. The model then is evaluated with published research and original data drawn from 19 cases of weapons decisions. The cases from which original data were drawn are highly representative of different services, mission areas, time frames, cost, technical complexity, and visibility.^ The findings challenge the empirical accuracy of subsystem theory in the area of weapons procurement and reveal the theoretical limitations of the model.
The Subsystem Model: Three PropositionsProposition 1: Political power (the ability to convert preferences into potiticat outcomes) is concentrated in the hands of a "tripartite alliance" ofmerr\bers of special interest groups, bureaucratic agencies or departments, and legislative committees or subcommittees whose interests reside in the same substantive policy area.