The challenge of accountability in complex regulatory networks: The case of the Deepwater Horizon oil spill

A puzzle that faces public administrators within regulatory governance networks is how to balance the need for democratic accountability while increasingly facing demands from elected officials to optimize oversight of industry by utilizing the expertise of the private sector in developing risk‐based standards for compliance. The shift from traditional command and control oversight to process oriented regulatory regimes has been most pronounced in highly complex industries, such as aviation and deepwater oil drilling, where the intricate and technical nature of operations necessitates risk‐based regulatory networks based largely on voluntary compliance with mutually agreed upon standards. The question addressed in this paper is how the shift to process oriented regimes affects the trade‐offs between democratic, market, and administrative accountability frames, and what factors determine the dominant accountability frame within the network. Using post‐incident document analysis, this paper provides a case study of regulatory oversight of the deepwater oil drilling industry prior to the explosion of the Deepwater Horizon rig in the Gulf of Mexico, to explore how the shift to a more networked risk‐based regulatory regime affects the trade‐offs and dominant accountability frames within the network. The results of this study indicate that a reliance on market‐based accountability mechanisms, along with the lack of a fully implemented process‐oriented regulatory regime, led to the largest oil spill in US history.     

College Guidance for All: A Randomized Experiment in Pre‐College Advising

Pre‐college advising programs exist in most disadvantaged high schools throughout the United States. These programs supplement traditional advising by high school guidance counselors and attempt to help underrepresented and disadvantaged students overcome the complexities of the postsecondary admission and financial aid processes. Existing evidence on these programs often uses within‐school randomization where spillovers and alternative supports may confound estimates. We provide the first evidence on a whole school intervention resulting from a school‐level randomized controlled trial in the United States. The college access program we study uses a near‐peer model where a recent college graduate works at the school assisting students in the application and enrollment process. Pooled results across the first three years of program implementation find no significant impacts on overall college enrollment. However, subgroup analyses reveal positive, significant effects among the groups most targeted by the intervention: Hispanic and low‐income students. Most of the impact comes through increasing two‐year college enrollment, but this appears to be new entrants rather than inducing students to move from four‐year to two‐year colleges. The observed positive effects for these subgroups attenuate over time. We attribute this drop in the estimated impact to departures in fidelity of the experiment. Even among the cohorts for which we find positive enrollment impacts, we find no significant impacts on college persistence.