No single risk framework will likely overcome all challenges, but embracing additional aspects can improve their usability in a world of certain uncertainty
Quantification and probability assessments can be great tools, but in using them, we need to be cognizant that when applied to systems with high levels of randomness, they may mask the variability and infuse a false sense of security, as has been argued in financial risk analysis. quantification infuses a false sense of security
different meaning of risk within different disciplinary fields
- Anthropocene risk
- Risks that emerge from/are related to anthropogenic changes, emerge through social-ecological connectivity, and exhibit cross-scale interactions.23
- Black swan risk (unknown unknowns)
- Risks of events of extreme rarity and severe impact. 24
- Disaster risk
- Combination of hazard, exposure, and vulnerability. 25,26
- Existential Risk and catastrophic risks, and cascading risks are extensions of this
- Engineering risk
- Combination of probability and consequences of event. 27
- Investment/portfolio risk
- Variability from historic baseline. 10
- Knightian risk
- Measurable uncertainty. Measurable meaning the ability to calculate the probability distribution of the adverse outcome.21
- Public health risk factors
- Factors (variables) correlated with higher incidence of adverse health outcome. 20
- Risk (de Moivre)
- Expected loss (probability multiplied by sum of loss). 10,19
- Systemic/networked risk
- Interconnected risks, often associated with high complexity and the potential for cascading risks. 28,29,30
- Toxicological risk
- Potential adverse health outcomes from exposure to hazardous/toxic agent.31
Risk and Resilience (connected but not really)
Risk and resilience are conceptually and practically connected, yet resilience-informed sustainability research has not engaged deeply with disciplinary risk approaches, or the field of risk science. On the other hand, risk science, when referring to resilience, generally refers to the engineering resilience definition, which does not account for complexity and assumes only singular stable states.15 Many traditional risk assessment tools were also created before insights regarding the complexity and intertwined nature of today’s world emerged, and consequently, they have not yet fully incorporated complexity ideas or been adjusted to tackle these.
When future outcomes are completely unknown, i.e., when we are not at all able to foresee what the impact of actions will be, and only know that the future is uncertain, then building resilience is a useful approach. Ecological resilience is defined as ‘the capacity of a system to absorb disturbance and reorganize while undergoing change so as to still retain essentially the same function, structure, identity, and feedbacks.’’
Risk in Finance and Black Swans
The field of finance, particularly in portfolio (investment) risk assessments, deviates from the above-described approaches to defining risk, by defining risk only as uncertainty or variance. 10,36 This approach is based on the assumption that uncertainty is measured against a known or fixed expected value.10 This is the common approach in financial (investment) risk analysis, where an expected value is set as some historical average and investments are analyzed on the basis of their variability and deviation (or risk) away from that baseline.37 The concept of ‘‘black swan’’ risks24 emerged as a critique against this conception of risk, particularly the use of Gaussian distributions and a historical average as a baseline for future risk. A ‘‘black swan’’ represents an anomalous event, and the critique centers on the premise that it only takes one anomalous event to break the belief in the past as the predictor of the future, such as the discovery black swans. In finance, a black swan is therefore a very rare event with a high impact.24
Different Kinds of Uncertainty
Uncertainty can be divided into aleatory uncertainty and epistemic uncertainty, where aleatory refers to uncertainty due to natural variability and epistemic uncertainty arises due to lack of knowledge.41–43 This distinction is most often used in the statistics and engineering communities, often related to the assessment or quantification of uncertainty. Aleatory uncertainty has to do with the inherent randomness (or stochasticity) of systems and the fact that the outcomes of the future are always hidden from us in the present.
Understanding risk to reduce uncertainty
We argue that epistemic uncertainty, the uncertainty that arises from a lack of knowledge, is affected by and a weakness of the siloed nature of risk and resilience approaches and could be reduced through an integrated understanding of risk.
True Uncertainty
Uncertainty also forms a big part of resilience scholarship, notably, true uncertainty, which is the space of complete unknowns.45 True uncertainty is similar to the upper levels of deep uncertainty (where outcomes are unknown) 44 and akin to the notion of fat-tailed risks or black swans within (specifically financial) risk scholarship
Uncertainty in calculability shouldn’t prevent mitigation action
In situations where uncertainty is conceived of as a limit in calculability rather than limits to our knowledge of potential outcomes, failing to address these uncertainties despite limited ability to precisely calculate their occurrence could lead to slower or unfocused mitigation of the causes. This could, in turn, translate into a concrete risk. One example is the Amazon rainforest. Although the global and connected impacts of the Amazon tipping from a rainforest to savannah are not precisely known, nor is the precise point at which it will tip known, there is increasing evidence confirming that deforestation of the Amazon will cause it to tip with escalated global warming as likely effects.53,54 As this is a known and undesirable outcome of the action of deforestation of the Amazon, not merely uncertainty about the future, it should therefore be accounted for in risk assessments and mitigation strategies, whether the probability of occurrence can be calculated or not. We argue that this more extensive definition of risk should guide how we assess and address sustainability risks today and in the future.
Challenges with Risk
Risk assessments need to incorporate variation across time, by estimating the randomness of the system, assessing the usefulness and accuracy of back casting and forecasting tools, and looking at which time horizons fully capture sustainability risks. Assessments also need to account for connectivity among risks to capture the intertwined nature of social and ecological dynamics, as well as the interaction of risks across scales. Lastly, risk assessment needs to be a continuous adaptive process where new knowledge is included, and risk is re-assessed and mitigated. As part of this process, it is also important to incorporate and acknowledge that risks can be, and often are, inequitably distributed due to power imbalances in who performs the assessment and who is affected.
meta
risk assessment frameworks #academic