The 10 costliest natural disasters in modern history have occurred since 1985, the year when the Wharton Risk Management and Decision Processes Center was launched by co-director Howard Kunreuther, a professor of operations, information and decisions. In late October, the 30th anniversary of the Center offered a timely opportunity to envision the future of risk management at an all-day conference at Wharton. “Thirty years is a long time in the history of any academic endeavor, but in this field the change has been truly remarkable,” noted University of Pennsylvania Provost Vincent Price, who opened the proceedings. “1985 was pre-Hurricane Andrew, pre-9/11 and pre-Fukushima, to name just a few of the more deadly and costly disasters,” Price said. “1985 was before the Internet and the cell phone — and before we understood the existential danger of climate change.”
Considering ‘Choice Architecture’
A major focus of the conference was on identifying strategies to overcome myopia and the “status quo bias” when making decisions about how to reduce long-term risk. Eric Johnson, co-director of the Center for the Decision Sciences at Columbia University, stressed the importance of talking about risk management decision making with real people who have actually made decisions. One such group includes those people responsible for building and managing today’s “incredibly complicated” physical infrastructure projects, which have wide-ranging effects on the environment.
Why are these infrastructure projects so complicated? He cited three reasons: First, in each of these projects, “there are many [different kinds of] stakeholders: politicians, taxpayers, bondholders, engineers. If we think of affecting those stakeholders … it gets to be a really complicated problem.” Second, “the time scale is enormous.” For example, “The train I took from New York to [Philadelphia] went through a tunnel that was engineered over 100 years ago. The curves that were so dangerous in North Philadelphia [where an Amtrak train derailed last spring] were engineered over 50 years ago. So someone who is designing [such an infrastructure project] now is thinking about a 100-year plan,” although they may not know it. Third, these massive infrastructure projects are “essentially irreversible. You could not move an airport [made years ago] because there has been a lot of infrastructure built around that” airport in subsequent years.
Armed with a full awareness of such complexities, the Institute for Sustainable Infrastructure, a not-for-profit education and research organization, has launched an important tool known as Envision, Johnson said. This planning and design guidance tool incorporates industry-wide sustainability metrics for all infrastructure types. Johnson explained that this tool tells stakeholders in these projects “what we are doing with this large system with respect to sustainability…. It has a great role. It becomes a checklist that people have to think about” with regard to planning for environmental risks.
“Some researchers have become interested in seeing how we can improve that kind of analysis by using ‘choice architecture,’” an approach that focuses on the way in which information is structured when it is presented to people, Johnson added. Decisions about choice architecture “have great significance” because they change the way people respond to the various choices that are presented to them, he noted.
To illustrate the importance of choice architecture, Johnson contrasted the way information about auto insurance policies was structured during the 1990s for drivers in New Jersey and Pennsylvania. In New Jersey, the default choice presented to drivers was “limited fault” insurance, while in Pennsylvania, the default choice was the “full tort” policy. Because of the way these choices were structured, “the difference in rates of adoption was astounding. Roughly 20% of drivers took the full tort in New Jersey, while 70% took the full tort in Pennsylvania. That’s because the box for the default choice was pre-checked. Roughly the same information was presented, but in a different way. All that was changed was what happened if you didn’t make a choice.”
Aware of such findings, risk management researchers have been thinking about how to make subtle changes in choice architecture that significantly improve the way that engineers make infrastructure-related decisions when they use the Envision system. “We changed the ‘zero’ mark in the scale from being the lowest [mark] to the second highest in the scale. We’ve been doing studies of students and engineers to see whether making this change will increase the level of care that they take with respect to rain water [for example]. The simple answer is ‘yes,’” he said. “The difference in the choice architecture will change what they do.”
Beyond Muddling Through
But will these sorts of incremental changes be sufficient to tackle the larger-scale risks resulting from global climate change? In his presentation, Granger Morgan, professor of engineering and public policy at Carnegie Mellon University, recalled, “More than a half-century ago, political scientist Charles Lindblom argued that ‘muddling through’ with incremental steps is frequently superior to trying to devise comprehensive policy solutions.” Now, “If climate policy is ultimately to be successful, it will have to be combined with some comprehensive policy.”
“If climate policy is ultimately to be successful, it will have to be combined with some comprehensive policy.” –Granger Morgan
To be sure, Morgan noted, “After decades of talk, the world is finally showing signs of muddling through toward serious policies to limit climate change. Promising events in the U.S. include the switch toward natural gas and the implementation of section 111 of the Clean Air Act. Internationally, the ongoing efforts in the EU, the agreement between the U.S. and China, and the efforts to limit emissions from aircraft and ocean shipping are also promising.”
However, while “muddling through provides a good start on climate policy, it is not good enough. Modest first steps that reduce greenhouse gases are wonderful, but to stabilize the climate, the world will have to reduce the emission of greenhouse gases by roughly an order of magnitude. It’s not too soon to think about how to avoid getting stuck with policies that don’t scale up, or how to avoid regulatory lock-in and move past early incremental steps to achieve much deeper reductions.”
Morgan provided five examples of the sorts of strategies “we ought to be thinking about.” First is adopting a total-system perspective on energy usage. Second, “developing strategies to support the continued operation of zero-carbon energy sources that would be cost-effective today if CO2 emissions were priced at anything like the social cost of carbon.” Third, “finding ways to preferentially favor technologies that can reduce emissions of carbon dioxide to very low levels or to zero.” Fourth, “applying more vigorous support to the development of future low-carbon energy sources and efficient energy systems.” And finally, focusing on technologies that can scale up.
The last strategy means “undertaking analysis today that focuses on identifying technologies that can scale up to meet longer-term needs, and avoiding pathways that would lead to future dead-ends,” he noted. “Muddling through may be the best strategy that can work today, but the community of stakeholders should begin to work now on technologies that can scale up to meet the more than 90% emissions reductions that we will need [in order] to stabilize the climate. Without some long-term visioning about how to move gracefully past short-term strategy, progress could stall…. The success of today should not be the burden of tomorrow.”
Knowledge ‘Co-production’ and ‘Social Trust’
Such an approach at long-term “visioning” will likely involve a very different way for assessing risks, noted Roger Kasperson, professor of geography at Clark University. Already, “We are in the midst of a major methodological change in the way we analyze hazards.” He explained that researchers are assessing “why the next 30 years are going to be completely different than the last 30 years, and [what are] some of the major difficulties we are going to experience in the process of getting there.”
“There will be much more attention placed on protecting vulnerable people.” –Roger Kasperson
According to Kasperson, “The risk analysis and hazard analysis we’ve relied on have principally [been based] on traditional science. Some people call this the deficit model. Gather your evidence and see what you can find out, with the notion of risk very often tied up with the notion of health risk or fatalities…. Traditional science has been primarily curiosity-driven. But sustainability — the new science that we are just beginning — is much more problem-driven. Traditional science tries to be value-free, while sustainable science is overtly value-centered. Traditional science uses a divide-and-conquer strategy, whereas sustainability science tends to be holistic. There has been little carry-over of traditional science…. Sustainability science is overtly policy-driven; the whole reason for doing it is tied up with policy change.”
Traditional science also tends to be expert-dominated, Kasperson added, whereas sustainability science “relies much more on co-production of knowledge. Traditional science relies, by and large, on the linear development of science, whereas in sustainability science, place-based analysis is becoming much more important. And capacity building and adaptation are major issues that are inevitably a major part of what the analysis should be.”
Finally, in traditional science, “Social trust is not really very relevant, but in sustainability science it is very central.” For sustainability science to take hold, “scientists have to be highly trusted…. We are at the very beginning stages of understanding what needs to be done when it comes to social trust.”
Although addressing long-term risk problems requires an approach that integrates findings from various disciplines, too much of the analysis to date has been founded on the findings of a single discipline, he argued. “We’ll have to rely much more on collaborative research and capacity building when we are looking for alternative futures…. There will be much more attention placed on protecting vulnerable people. Vulnerability did not become a major part of risk analysis until 10 years ago…. The whole role that vulnerability plays is still at a relatively early stage of analysis.”