Interorganizational Policy Implementation: A Theoretical Perspective

During the last decade or so, there has been a virtual explosion of research on the subject of implementation. Some scholars have produced case studies of implementation success or failure in various policy sectors. I Others have aimed at producing a body of implementation theory. Of these, some have been almost exclusively theoretical, while other, mostly later, work has attempted some synthesis and integration of empirical findings into a set of general propositions.3 There have been methodological advances, comparative analyses, and studies of the effects of federal structure on implementation.4 Some analysts have tried to suggest that different categories of policy may have predictably different implementation results, while others have examined the implementability of judicial policies.5 One group of researchers has explicitly adopted a topdown perspective and has sought to find ways of structuring mandates so that implementers will have little chance to exercise discretion.6 Others, however, either have raised general criticisms of such an approach or at least have suggested conditions under which successful implementation would require substantial delegation.' Considerable progress has been achieved in carving out this research field and in developing useful knowledge on the topic of implementation; however, much remains to be done. There is, as yet, no general agreement on a predictive theory of implementation or even on what variables are most important to consider. This paper is an attempt to systematize our knowledge in an important class of implementation efforts, interorganizational implementation. Often mandates require the efforts of two or more agencies for implementation. There are today many patterns of sustained cooperation across agency lines and levels of government, a phenomenon which the growing field of interorganizational theory addresses. How do such patterns begin? Some were obviously responses to policies that