Implementation Structures: The Use of Top-Down and Bottom-Up Approaches to Policy Implementation

Implementation research emerged as an effort to understand the “missing link” between the expression of a governmental intention and the world of action and results. In many policy settings, this requires implementation structures or networks comprised of parts of organizations both within (vertically) and across (horizontally) levels of government. Increasingly, this involves structures that incorporate organizations from the private and nonprofit sectors. Therefore, effective interorganizational policy implementation requires building networks with the correct balance of federal, state, and local control to achieve the collective objectives of these actors. Consequently, the challenge of managing within these networked implementation structures is quite different than what is found in a typical hierarchical organization. There are three stages of intellectual development in implementation research. Early scholarship typically used case studies to examine detailed episodes of policy implementation to identify problems and challenges. A more sophisticated approach to research soon emerged that emphasized identifying variables crucial to implementation “success.” Two competing perspectives soon characterized this stage of intellectual development. The top-down approach argued that implementation problems are minimized through careful specification of procedures. From their perspective, implementation was largely an administrative challenge. Conversely, the bottom-up perspective argued that effective implementation allows policy to be adapted based on the interaction of a policy with the local institutional setting. For bottom-uppers, context matters, and implementation involves bargaining rather than the explicit control of higher-level decision makers. Some of the notable efforts to synthesize these perspectives are then examined. However, these efforts were hindered by obstacles such as different philosophical perspectives and pragmatic realities about how a polycentric governmental system functions, the failure to embrace a longitudinal perspective, and the improper specification of the unit of analysis. While the volume of research has declined since its heyday in the 1980s, the so-called “third generation” of research that succeeded it has become much more rigorous. However, a generally agreed upon theory of implementation remains lacking. A competing approach to implementation scholarship emerged during the top-down and bottom-up debate. It argued that the choice between these two approaches was a false one. Instead, the core implementation challenge is one of governance and crafting implementation structures that deliver services. This stream of research grew largely out of the bottom-up approach but argued that the proper unit of analysis is the “network” rather than a policy or statute. This proved to be a useful methodological approach for identifying the networks used to “implement” policies and programs. A variety of new perspectives on “networked” policy implementation soon emerged out of this implementation structure tradition. Research on implementation networks was soon joined by the growth of new literatures in areas such as intergovernmental management (IGM), network governance, collaboration, and institutional analysis and development that also provide useful insights about the challenge of “managing” within implementation structures. Moving forward, there is much that implementation scholars can draw upon and contribute to advance the collective understanding of how to “manage” within networks.