Volume 19, Number 6
Robert Klassen's research shows that managers don't pay enough attention to operational failures in other firms.
The explosion and resulting oil spill on British Petroleum's drilling platform in the Gulf of Mexico in 2010 captured our attention for months. CEO Tony Hayward characterized this as a "low-probability event," and observed that "BP has learned a tremendous amount…that can enhance safety in the future." Yet, others noted that the spill was not an isolated event. In the years leading up to it, other firms experienced similar, but smaller scale problems. So, why had BP not learned from others' earlier mistakes?
Experiences like this suggest that managers could learn more, much sooner, from failures in other firms. In one stream of his research, Ivey Professor Robert Klassen looks at how managers think about risk. Recently he and Georgia Institute of Technology Professor Manpreet Hora, an Ivey PhD grad, published a study in the Journal of Operations Management. Existing research tends to focus on helping managers to replicate the success of other firms, not avoid their failures. For example, benchmarking is very common. "In this study we were trying to understand why managers dismiss problems that occur in the operations of other firms, and more specifically, when observing managers are more likely to dig deeper to see what their own firms could learn," says Klassen. "In doing so, we can start to identify options to overcome barriers to learn from others' failures."
To conduct their study, Klassen and Hora recruited 182 experienced risk managers from both the chemical and financial services industries to participate in a vignette-based experiment. These two industries, although quite different from each other, were chosen because they both had developed risk management systems.
In the experiment, participants answered questions on scenarios that illustrated relatively rare, but high-impact failures, such as a plant explosion or a client related large-scale information technology failure. In each scenario, the participant was asked to assume the role of manager in an "observing firm," who observed the failure in an "incident firm." The experiment explored two questions. First, is the incident firm a market leader? Second, how important is it for observing and incident firms to have the same operational processes? After manipulating these factors, the researchers asked participants to indicate how likely they were to gather additional information about the failure, and then disseminate any learning in their own firm.
The researchers found that having virtually identical operational processes was the most significant factor, with market leadership having less, but still important influence. Even if the operational processes between the observing firm (e.g., agrochemicals) and incident firm (e.g., polymers) differed marginally, the likelihood to gather information dropped significantly. "We were really struck by how little a difference it took to move a firm right off a risk manager's radar," says Klassen. "We saw a lot of heavy filtering going on." However, if the incident firm was a market leader, that overcame the filtering.
Interestingly, the participants in both the chemical and financial services industries were consistent in the way they thought about risk and learning from failure, although risk managers in the chemical industry were somewhat more proactive about information gathering. The researchers also found that the educational level of a risk manager played a significant role. Risk managers with a graduate education were more likely to seek out further information that might help their firm learn from the other's experience.
Klassen says that this research suggests two important lessons for managers. Firstly, firms should develop an explicit strategy about how they gather information on operational failures at other firms. "At a very senior level, firms need to identify more systematic ways of learning from firms in closely related industries that use somewhat similar operational processes," he says.
Secondly, firms might be able to make better use of industry associations to gather information about failures and facilitate learning. For example, the Responsible Care Program in the chemical industry was set up to improve health, safety, and environmental performance, and has been adopted by many in the industry. "Industry associations need to really stress the message that we want to learn from the mistakes of others to protect the whole industry," says Klassen.
Alternatively, regulatory agencies might serve the public good through the formation of an industry watchdog. An example is the U.S. Federal Aviation Administration (FAA), which takes great pains to investigate any failures that jeopardize aviation safety. Operating practices can then be changed for the entire industry.
Klassen would prefer to see firms and industries, rather than regulatory agencies, working together to improve risk management. For example, the learning can occur more quickly. But he believes managers still have a long way to go. "Managers don't pay much attention to the problems experienced by others when they have even slightly different processes," he says. "It's going to require some major educational efforts to reshape risk managers' perceptions."