David B. Spence


In the last decade, positive political theory (PPT) models of agency policy-making have emphasized the tools politicians can use to control or influence agency decisions. These models are, in part, a reaction to earlier economic and other models of agency policy-making which emphasized the agency losses attendant to the delegation of decision-making authority to agencies by politicians. The more recent PPT works contend that politicians use the tools of ex post and ex ante control to overcome some of the agency problems associated with delegation (such as the inability to foresee the issues the agency will face), in part by enlisting interest groups in the battle to control agencies. These recent PPT models of political control do a good job of illustrating how and why politicians try to influence agency policymaking; but they overstate politicians' ability to do so, for two reasons. First, commonly-employed methodological assumptions in positive models tend to obscure the most important impediments to political control. Second, the antecedents to the current PPT literature posed a false dichotomy between agency autonomy and good government, one which some positive theorists seem to continue to accept, at least implicitly. This article examines these positive and normative biases in the PPT literature on the political control of agencies, and argues that PPT policy models which abandon these assumptions will do a better job of (1) describing the agency policy-making process and (2) accommodating the fact of agency autonomy in the policy process.

Included in

Law Commons