Association for Public Policy Analysis & Management

Historically, government agencies were vested with authority to manage specific policy agendas that often fell uniquely in their jurisdiction. Over time, jurisdictions began to touch, overlap, and clash as policies grew more complex, which often spawned ‘turf battles,’ introduced differing legal interpretations of roles, and produced policy blind spots. Today, policymakers face complex mazes of problems and options; it is often difficult to engage all relevant constituencies (or know who they are), fully anticipate second and third order effects of decisions, and operate in policy or legal frameworks that feel rigid or antiquated. It also reveals a fundamental truth for modern policymakers – no agency has a holistic view of what needs to be done or is sufficiently empowered to optimize policy implementation. Moreover, their ability to collaborate is often constrained by perceptions or lack of knowledge of the authorities, roles, and responsibilities of their agency peers.

We developed a Policy Mapping method to visualize the relationships and interdependencies within and across policy environments to overcome these constraints. We systematically extract and classify relevant policies; identify potential areas of friction or collaboration; and produce visuals. We used this method at multiple Federal agencies, and the results from our analyses have been presented and used by agency leaders in areas ranging from international assistance to cyber security. The methodology is flexible enough to address specific issues such as medical countermeasures, and understanding the impact of broad legislation (e.g., tax policy implications of the ACA), to broad, interagency policy challenges. Major issue areas today, such as climate change and the next era of health reform, are not siloed; they are felt by multiple actors. Within the current interdependent policy ecosystem, a methodology that applies a broad view, with an adaptable breadth and depth, positions policymakers for identifying the first place to act with defensibility.

Over the past decade, we adapted our method by levering advances in analytic tools and techniques and adding automation. The goal was to transform our outputs from a decade ago to a more dynamic, predictive approach to help policymakers anticipate, plan for, and respond to shifts in the policy ecosystem. Our five-step Policy Mapping method: (1) inventories the policy ecosystem by scoping the depth of policy to be examined; (2) catalogues relevant policies; (3) creates visuals to map the key actors, functions, and relationships; (4) identifies areas of friction from policy and operational perspectives; and (5) formulates a roadmap of issues and opportunities based on the map and analysis. To advance the method, we integrated Natural Language Processing and Supervised Machine Learning; we created a comprehensive data lake of policies; and enhanced the utility of the data by employing techniques such as semantic analysis, document similarity, and multiclass classifications. This process of digitizing policies enables us to dynamically visualize relationships, policy impacts, and other relevant data (e.g., budget) in real time. Ultimately, our methods help policymakers move beyond laborious, manual activities that produce static analyses to a dynamic, visualized, and intuitive approach to make tough policy choices.