Association for Public Policy Analysis & Management

A fundamental challenge for public managers and policymakers is diagnosing the nature and sources of conflicts that can shape policy decisions and the capacity to effectively implement those decisions. The Policy Conflict Framework (PCF) aims at understanding the sources and outcomes of policy conflicts in the policy process (Heikkila and Weible 2016). PCF organizes inquiry of policy conflicts around a common language, labeling and categorizing key concepts, and provides a platform for theoretically guided research. The goal of the PCF is to analyze and compare episodes of policy conflicts over time, including their setting, the characteristics of actors involved, and outcomes, and to recommend strategies for dealing with policy conflicts. While the PCF has been used to assess policy conflicts in a particular location, it has not been employed to examine characteristics of policy conflicts across policy contexts or over time. Such comparisons are essential for understanding the intensity of policy conflicts, and ultimately what factors drive different types of conflicts.

Unconventional oil and gas development is a prime example of the types of policy issue where conflicts among diverse types of actors are engaged at the state level in shaping policy debates. Guided by the PCF, this study seeks to address the following related research questions: How do policy actors involved in unconventional oil and gas development policy networks compare across states? Is their involvement stable over time? Do characteristics at the state level, such as oil and gas production capacity, explain differences in the typology of these networks? We hypothesize that the networks vary significantly across states and over time, based on both systemic and individual-level attributes.

To address our research questions, we collect and analyze news reports on unconventional oil and gas production published in five different states that vary in their levels of oil and gas activity (California, Colorado, Montana, Ohio and Pennsylvania) from 2008 to 2017. We develop a well-documented codebook for analyzing articles in the two newspapers in each state, focusing on information about policy actors, their beliefs about the topic, and potential conflictive views with others. In additional to manual coding of the articles, we use machine learning based techniques to perform our analysis. In the initial analysis of newspaper articles in Colorado and Ohio, we identify around 1300 and 500 articles respectively for each state, with Colorado having more policy actors, more activities and more intense level of policy conflicts than Ohio. We expect the instruments developed in this study will allow for efficient and consistent data collection efforts on policy conflicts and policy networks over the long run.

In addition to building knowledge about the networks of actors involved in policy conflicts over oil and gas development, this research is among a small group of studies thatexamine the evolution of policy subsystems and policy networks over at least a decade across different states in the U.S.