Systems Resilience in the Context of Covid-19

Reported by CSaP Policy Intern Julia Amtmann

Resilience concepts can be used to analyse entire systems and their ability to respond to shocks and disruptions. The Covid-19 pandemic has brought to light the urgent need to build resilient systems to tackle pressing policy problems.

In late October, CSaP Policy Fellows gathered online with Professor Frank Kelly and Dr Magda Osman for the first in a series of Resilience Seminars to reflect on the experience of doing public policy during these turbulent times.

Frank Kelly, a former Chief Scientific Advisor to the Department for Transport, and Professor of Mathematics of Systems at the University of Cambridge, gave critical insight into the performance and behavior of telecommunications networks as an example of how systems and resilience are connected. When decentralized subsystems are being improved to make a part of the network of the system behave more reliably, other subsystems are often forced to compensate for fluctuations in stress levels on the system as a whole. When failure occurs, it's more pervasive, more widespread, and probably has larger consequences as the effects on the system are aggregated. Dr. Frank Kelly suggested forecasting systems’ stress levels through predictive models in order to enhance the resilience of the whole system.

Magda Osman, reader in experimental psychology at Queen Mary University of London, led the conversation on individual and social resilience. Individual resilience describes an individual's capacity to find opportunities in tragedy and turn adversity to advantage. In contrast, social resilience refers to the capability to foster, engage and sustain positive relationships and to recover from and grow as a result of stresses. She emphasized that there are potential costs attached to the assumptions one makes about an individual's or system's capacity to adapt to externalities or shocks. One such assumption is that a system is resilient enough to overcome and learn from these shocks and, in turn, becomes even stronger. Alternatively, another common assumption is that a system lacks the resources to cope with or respond well enough to the shocks that occur, which can lead to overregulation. This shows the importance of designing interventions based on an accurate understanding of the system’s resilience, as misestimations of its resilience can have negative consequences.

During the discussion, Policy Fellows debated the importance of a systems memory and how it relates to both the system’s goals and resilience. The memory helps one to set goals as previous experiences might just fluctuate fractionally to a current situation. To hearken back to former shocks can thus strengthen resilience. But there might be a cost inherent as the memory might not be appropriate for a new shock or disruption that occurs. In this case trying to adjust for the new process can send one in the wrong direction. Participants also explored the problem of coping with chaos and complexity and how the intricacy of networks and the instant feedback loops can generate conditions for chaos. A systems ‘pain’ was essential in the discussion around chaos caused by the Covid-19 pandemic. The group agreed that models need to be found which not only react to hospital beds filling up, but which also predict future incidents and set precautions in place accordingly. Hence, the ‘pain’ in systems cannot arise and a problem’s complexity decreases. The solution seems evident: systems need to be so strong that upcoming problems can be managed. For this, one needs to think beyond, not just at the next problem. A shared learning experience over time is needed and these policy fellow seminars are helping to funnel the progress towards systems with resilience.