On the Edge: Heeding the Warnings of Unusual Events

When organizations pay inadequate attention to unusual events, the possibility of breaching a safety barrier increases. With hindsight it often appears that full advantage is not taken of what is known. Part of the reason organizations neglect apparent warnings is because of limited resources and the way resources are allocated. Drawing on concepts from the resource-based view (RBV) of the firm, this paper examines situations where resource availability and use can compromise safety by initiating drift. Empirical evidence from the nuclear power industry is reviewed to demonstrate the relationship between resources, resource deployment, and drift. Drift is influenced by error signals, feedback loops, and imperfect watchfulness. Resource availability and use can initiate drift, but they do not necessarily yield catastrophe. Should organizations fall into a threatened position, they can enlist characteristics, behaviors, and capabilities to stabilize the situation, avoid breaching the safety border, and achieve greater security. Through an in-depth study of two plants with contrasting reputations for safety, this paper identifies the characteristics, behaviors, and capabilities that organizations on the edge exhibit. The characteristics, behaviors, and capabilities of the plant with the stronger safety reputation changed after a significant reduction of resources. Resource reduction led to movement toward the safety border, which in turn led to change. Evidence shows that organizations operating closer to the border of safety respond to warnings from unusual events in a resilient rather than anticipatory way. To overcome unexpected problems and limitations, they rely on after-the-fact intervention rather than preparation, foresight, and the provision of countermeasures. Once they recognize they have problems, they explore for novel approaches, typically seeking new knowledge and skills by dismissing old staff and hiring new people, buying advice from outside experts or consultants, and modeling, benchmarking, and copying other organizations' best practices. Organizations on the edge initiate change in response to external demands for change often only after painful public incidents. These organizations emphasize hierarchy and powerful headquarters staff, and their leaders tend to act as commanders and controllers rather than as catalysts and facilitators. The focus of this paper is not on the occurrence of accidents per se, nor on highly reliable operations, but on the precursors and consequences of drift within a safety border. Operating on the edge of the border extends beyond safety to other performance measures and beyond nuclear power to other industries. In settings as diverse as U.S. banks, hospitals, and universities, Israeli kibbutzim, Japanese kereitsu, and Korean chaebol, operating on the edge has become more common.