Already a fairly large body of literature has developed using Niskanen's theory as a starting point, showing how changes in Niskanen's assumptions can lead to a more realistic, yet still rigorous theory of bureaucratic behavior. This paper will review some of the more important elaborations of Niskanen's theory to date and suggest a still more fundamental alteration of the assumption set. Basically, this paper will argue that the bureau head's supply decision cannot be regarded as an isolated decision, constrained by a fixed budget-output schedule. Rather, the bureau head's supply decision and the sponsor's budget appropriation decision must be analyzed in strategic interaction. This perspective suggests a game-theoretic model and permits the analyst to ask questions about the uniqueness and efficiency of joint solutions that are difficult or impossible to investigate with simpler models.

The author, an assistant professor at California Institute of Technology, wishes to express his appreciation to Joe Oppenheimer and Norman Frohlich for their helpful advice and encouragement during work on early drafts of this paper, and to William Niskanen for his thoughtful critique and suggestions. Published as Miller, Gary J. "Bureaucratic compliance as a game on the unit square." Public Choice 29.1 (1977): 37-51.

Submitted - sswp141.pdf