413 Hofferbert in Transit: A Dynamic Stages Model of the Urban Policy Process Robert J. Eger, III and Justin Marlowe University of Kansas Abstract This article presents a reformulation and empirical test of Hofferbert’s (1974) “stages” heuristic, a model that fell largely out of favor due to criticisms that it does not effectively account for (1) feedback loops, or the possibility that policymakers learn from past experiences, and (2) the importance of intergovern- mental relations. We update this model and apply it in the context of urban transit policymaking, using revenue flows to and from the government as an indicator of both recursive behavior and intergovern- mental influence. The results suggest that these modifications reestablish the explanatory power of the stages model. Introduction The process by which public policy is made, implemented, and ultimately modi- fied is a timeless issue in policy studies and public administration. To date, research examining that process has focused primarily on the complex interactions among actors, institutions, and other factors that bring about observable patterns in public policy outcomes. The inherent complexity of those interactions has led policy process scholars to place great value in heuristic frameworks that provide stream- lined explanations for which factors matter and why (Blomquist, 1999; Brunner, 1991; Nakamura, 1987; Sabatier, 1988; Sabatier & Jenkins-Smith, 1993; Smith, 2004; Williams, 2004). Much of the development and debate within the policy process literature stems from disagreements over the elements, dynamics, and com- parative value of these competing frameworks. This article revisits one of the oldest and perhaps the most important debates within the policy process literature by reconsidering and modifying Hofferbert’s (1974) often-cited “stages heuristic.” Although it was tremendously influential for several decades, the Hofferbert model fell largely out of favor during the 1990s in light of two key criticisms. First, as Blomquist (1999) and others have argued, its nonrecursive nature renders it unable to account for “feedback loops,” or the potential that policymakers might learn from past experience. As a result, it fails to explain the observed incremental changes in policy over time. A second criti- cism, often attributed to Sabatier and Jenkins-Smith (1993), is that it does not adequately account for the importance of intergovernmental relations. These criti- cisms provided the catalyst for the development of several alternative frameworks, many of which comprise the core of the contemporary policy process literature (for example, see deLeon, 1999; Sabatier & Jenkins-Smith, 1993; Ostrom, 1999). Although much has been learned from these alternative accounts, this article shows that the stages model may have been prematurely abandoned. It does so by reformulating this model to include both the dynamics resulting from intergov- ernmental lobbying, as well as an explanation for how relationships across the Review of Policy Research, Volume 23, Number 2 (2006) © 2006 by The Policy Studies Organization. All rights reserved.