Consensus Building Approach by Larry Susskind
Have you ever noticed that all the writing about risk perception and risk assessment assumes a unitary mind? One single decision-maker is always trying to figure out what to do in the face of some potential hazard — a possible earthquake, tornado, hurricane, sea level rise, terrorist plot, and something equally terrible. In actuality, most risk management choices involve collective decisions. Whether our community decides to build a sea wall to protect coastal properties from storm surges isn’t up to one person. We need to make a collective choice.
Societal risk (R) is calculated as follows. First, figure out the probability (p) of a hazard occurring (H). Then, multiple that by the impact (I) the hazard will cause if it does occur. R = p (H) x I. If there is a 1% chance of a flood occurring in our community in a give year, and major (100 year) floods in the past have done a $10 million in damage, then the annual societal risk associated with a 100 year flood is $100,000 a year. Easy, right? Well, first we have to be able to forecast the likelihood that a major
flood will occur in a given year. Climate change is making that hard to do. Whatever the past pattern of flooding has been, it may be changing. And, it is not easy to calculate the impacts of a major flood.
The $10 million figure is probably just a measure of average property damage in the past. What about the long-term impacts on the natural environment and all the ecosystem services that will be lost if soil or wetlands are washed away? And, there may be loss of life or injuries involved. How should we put a price tag on these losses? And, the psychological and emotional damage involved in losing one’s home or neighborhood is not insignificant. If our community has to decide whether to build a $1 million sea wall to protect some, but certainly not all, coastal property, which costs and which benefits to whom have to be assessed and priorities have to be set.
The more uncertainty attached to an estimate of societal risk (in terms of both p (H) and I), the more difficult it is to reach agreement in a community about the measures that should be taken. Risk management can sometimes reduce the probability of an event happening. Most of the time, though, risk management measures are aimed at reduce impacts. So, if we have an emergency evacuation plan, and we practice it ahead of time, we can avoid some of the costs (i.e., deaths and injury) when the hazard occurs. There’s a cost to preparing and rehearsing. And, not everyone will want to pay the costs involved. If the “best” risk management approach involves restricting what is allowed to be built near the shore, or imposing new construction methods, other groups are likely to object. Some will oppose such measures because they will experience additional costs if those measures are implemented. Others will object in principal, arguing, for example, that individuals have a right to build where they want and the way they want. Of course, the costs to the rest of the community (of having to provide emergency evacuations to those who refuse to move, of needing to build and maintain roads and other infrastructure for the benefit of the people who refuse to move, of facing higher town wide insurance premiums and bond rates) are rarely taken into account in these conversations.
So, risk management involves collective choices, but the theory and practice of risk management always seem to assume a single (rational) individual will weigh all the costs and benefits and do what’s best. When there are multiple stakeholders, and they stand to gain and lose different amounts, and they hold different views on what ought to be an individual’s responsibility and what should be the government’s responsibility, risk management decisions become extremely contentious.
One approach to dealing with these difficulties is to rely heavily on technical or scientific analysis. If there were a way to produce an indisputable forecast of the likelihood of H occurring and the likely impacts if it does occur, there would still be differences among stakeholders with regard to the actions that should be taken, but at least everyone would be working with the same forecast. As it turns out, the “systems” we need to model are quite complex. And, even if we add a lot more computing power, the complexity and uncertainty involved in most socio-ecological system interactions makes forecasting extremely difficult. This means that the task of reaching a collective judgement about deciding how best to handle R is doubly hard. We can’t get a definitive forecast AND we have different views about the appropriate response to whatever forecast is available. As climate change adds even more uncertainty, the job of managing the risks associated with sudden climate change becomes even more difficult.
Another approach, one that de-emphasizes the search for a definitive technical forecast, is called scenario planning. This is a technique that begins with a wide range of possible forecasts (each the result of starting with a different driver of change). The task of risk management involves searching for “no regrets” actions that will reduce societal risk whichever “future” materializes. So, for example, instead of building an expensive sea wall that protects only a few coastal properties, a community might decide to build up natural barriers along the shore that provide some additional protection to all landowners AND enhance ecosystem functions at the same time (whether or not there is a terrible storm). While this may not protect a particular property owner as well as a sea wall if a ferocious storm occurs, it will provide a range of community benefits every year that more than justify the collective investment.
Scenario planning assumes we can not produce a reliable forecast of societal risk. So, like a financial investor trying to figure out where to park their money for the long term, a community is better off with a diversified portfolio rather than a single bet. Scenario planning provides a wide range of forecasts, each based on a different set of assumptions about p and I. The can use the full set to bracket the risks it probably faces. Then, they can ask which risk management strategy makes the most sense given the bracketed range of possible futures.
To make this work in practice, three things have to happen. A broad set of stakeholder representatives in the community need to engage in scenario planning. If they aren’t involved directly, they are not likely to have much confidence in the multiple forecasts that emerge. Also, they will probably need technical assistance to sort through the various risk reduction options and their likely costs associated with each scenario. Second, the product of the scenario planning effort ought to be widely publicized. Risk management is not something that should be turned over to our elected officials to decide on a community’s behalf behind closed doors. The public needs to be heard once it has reliable information. There is no correct decision only a plausible decision given all the concerns that the various stakeholders bring to the conversation. Third, every community should probably put a collaborative adaptive management (CAM) strategy in place. When we are dealing with complex and uncertain systems, every move we make should be seen a provisional. So, once we’ve made a move, we ought to monitor the results closely (as well as the results in other places where they have taken a different approach) so we can make adjustments. Risk management is a continuous process.
Most people haven’t stopped to think about the chances of an earthquake occurring in the place where they live. Yet, at some point, some people decided whether or not to impose construction standards that can reduce the risk of death dramatically. If buildings have to be earthquake “proof,” they are probably going to be more expensive. Even if we have very little confidence in our ability to predict the likelihood of an earthquake occurring (pH) or of estimating the impacts (I) if it does occur, we are probably glad to be living in a earthquake-resistant building. I assume that the same thing is true about climate change. Even if we can’t predict sea level rise or increasing storm intensity (pH) with much confidence, we have an idea that the impacts (I) would be enormous. If there are ways a community can increase its resilience and reduce its vulnerability (and achieve a variety of other goals at the same time), the collective choice should be relatively straight-forward. Of course, this will only work if the community has access to the results of a transparent scenario planning process.
In a risky and uncertain world, we should treat risk management decisions as collective choices. We should not pretend that we can forecast p(H) or I very accurately or that there are “correct” risk management decisions. Instead, we should commit to a transparent process of sustainability planning that yields the information stakeholders need to participate in collective choices. And, we should admit that whatever actions we take are, at best, provisional, while we learn more and make adjustments.