Can systemic risks ever be effectively governed?


Marc Gordon

Scott Williams

United Nations Office for Disaster Risk Reduction

This is the fifth in a series of eight articles co-authored by Marc Gordon (@Marc4D_risk), UNDRR and Scott Williams (@Scott42195), building off the chapter on ‘Systemic Risk, the Sendai Framework and the 2030 Agenda’ included in the Global Assessment Report on Disaster Risk Reduction 2019. These articles explore the systemic nature of risk made visible by the COVID-19 global pandemic, what needs to change and how we can make the paradigm shift from managing disasters to managing risks. 

Governance generally refers to actions, processes, traditions and institutions (formal and informal) to reach and implement collective decisions. Risk governance is “the totality of actors, rules, conventions, processes and mechanisms concerned with how relevant risk information is collected, analysed and communicated and how management decisions are taken.” Risk governance is usually associated with the question of how to enable societies to benefit from change, so-called “upside risk”, or opportunity, while also minimizing downside risk, or losses. In contrast, systemic risk is usually seen as downside risk.

As illustrated by COVID-19, the realization of systemic risk by definition leads to a breakdown, or at least a major dysfunction, of the system as a whole. Assessing, communicating and managing – in short, governing – systemic risk is compounded by the potential for losses to cascade across interconnected socioeconomic systems. Losses can cross political borders (including municipal and national boundaries), can irreversibly breach system boundaries and can impose intolerable burdens on entire countries. Systemic risk governance is also confounded by almost intractable difficulties in identifying causal agents and in assigning or attributing liability.

What needs to be set up so that institutions can govern systemic risk? Like any emerging phenomena, systemic risk cannot be measured by quantifying each of the contributing parts. This means that effective governance must consider the interconnected elements and interdependencies among individual risks, within and across systems.

For this purpose, a network perspective, with attention to interconnected nodes or agents, is useful. Individual and institutional decision makers also need greater accountability and responsibility, for example, through the establishment of the principle of collective responsibility.

Systemic risk governance requires new institutional structures. This was recognized after the global financial crisis in 2008. Prior to that, early warning systems (EWSs) were in place to identify precursor signals and anomalies in the overall performance of the complex financial system. Yet they failed to detect what are now understood to be clear signals. In 2007, the estimated probability of a financial crisis occurring was between 0.6% and 3.4%.

Financial systems operate in a siloed fashion. Constituents operate from their perspective and within their mandates. Yet such systems often become corrupted. Or they behave in a way that is suboptimal or pro-cyclical at a systems level, thus reinforcing underlying dynamics. Few organizations have the wherewithal to investigate at a system level, let alone a system-of-systems level. Consequently, ownership of the problem is often lost.

The global financial crisis prompted the development of new – or the reshaping of old – institutions and mechanisms to identify, and ideally prevent, future systemic risks in the financial system. But, post-crisis governance structures remain insufficient to prevent a further financial crisis – or the realisation of other systemic risks, such as the current COVID-19 pandemic.

The financial crisis focused attention on global interdependencies and cascading risks with potentially catastrophic consequences. But there are a worrying number of other potential triggers. These include, amongst others, extreme climate events, armed conflict, forced migration, food system disruptions, food and water shortages, unregulated digitalization, loss of biodiversity and zoonotic pandemics such as COVID-19. The climate crisis is a systemic risk with potentially catastrophic impacts cascading through financial, ecological and social systems. Climate change also has one of the most developed global governance regimes.

Neither the governance of the financial system nor the climate system can claim full success. But both have raised awareness of the necessity, and spatio-temporal complexity, of governance regimes to address systemic risks at the global scale. Moreover, the financial and climate governance regimes have brought attention to the complex web of challenges. One major challenge is establishing causal attribution of systemic losses as the basis for assigning accountabilities and responsibilities. This is essential for risk governance.

Attribution in relation to systemic risk is generally unclear, in particular where large uncertainties exist in determining the causal effects across complex geospatial regions, across stakeholders, and across sectors. For example, experts generally agree climate change amplifies the risk of extreme droughts and floods in some regions. Yet attributing losses from any event to human-induced climate change is still unachievable. As we observe in the COVID-19 pandemic, attribution is further complicated as systemic risk can evolve up to the global macroscopic scale, through disruptions at the microscopic scale; so-called “scale-free properties”, or through behaviour that is not directly linked to the disruption it causes in a specific system.

So, the difficulty of attributing accountability bounds the solution space for the reduction of systemic risks. It also hampers the urgent development of a joint vision defining clear approaches to management and the development of much needed policy responses at appropriate scales.

Another challenge, although not unique to systemic risk, is the often deep uncertainty surrounding the triggers, exposure and cascading consequences. Adopting a systems approach that takes account of network dynamics and social processes can form a basis for designing risk governance approaches in this context.

Beyond uncertainty, the lack of understanding of the systemic nature of many risk contexts poses a more daunting challenge. One suggestion taken from the climate risk community is to use a triple-loop learning process. from reacting to reframing and finally to transformation. This is also in line with suggestions made towards an adaptive risk management framework with a focus on solutions with multiple benefits.

GAR2019 graphic

The need for inclusive stakeholder expert processes is at the core of any risk governance framework, including systemic risk governance. These are important for co-designing and co-generating information, evidence and responses or solutions. While the importance of stakeholder buy-in has become clear, there are special challenges for systemic risks. For one, the cascading and uncertain nature of the losses means that stakeholder communities are ill-defined and often span political borders. Because of the uncertainty, the issues are characterized by varied perspectives on the nature of the problem and its solution, as well as different “risk constructs” on the part of the stakeholder communities.

For the “realists”, the risks can be objectively assessed in terms of their likelihood and impact. Whereas for the “constructivists”, the existence and nature of risk derives from its political, historical and social context. That is, it is constructed.

The two divergent views can have a significant impact for policy implementation. Modernity reflexively relies on increasing complexity to manage the very risks it creates. These in turn cause disasters that are often embedded in the construction of social organizations and institutions. Consequently, iterative approaches are better able to determine potential conflicts and possible solutions by identifying precursor signals or anomalies in system performance at the earliest possible moment.

Human agency may play a less-important role in some systemic risk considerations (for example, in supply chain risks) than in others (for example, a pandemic like COVID-19). This is important to consider for the corresponding governance approaches. The question is related to the optimal complexity to govern systemic risk. That is, how detailed the approach should be, given that there are always limited resources.

In the case of complex systems and systemic risks, current measures and approaches represent a collection of failed attempts. Nevertheless, the approaches are raising awareness and addressing challenges. These can shed light onto critical aspects of what is itself a complex issue – systemic risk governance.

Emerging approaches (for example, the International Risk Governance Center (IRGC) systemic risk governance guidelines) seek to address the difficult problem of assessing or measuring systemic risk, of modelling cascading consequences, of applying different management instruments, and of implementing participatory processes.

Successful implementation of such systemic risk governance approaches assumes flexibility and (continuous) adaptation to context (that is, adopting an iterative process). It is contingent upon strong leadership (with mid- to long-term focus), to prove the willingness to adapt or revise often non-linear, non-sequential processes, and the willingness to accept and resolve trade-offs.

Applying insights from more conventional risk analysis, risk communication and risk management to connect systemic risk with more traditional risk governance approaches can speed up the transition from managing disasters to managing risks.

The next article in the series (#6 of 8) builds off this exploration of some of the necessary elements to consider for systemic risk governance. It focuses on the importance of building collective intelligence to understand how parts of systems are related. It also explores the implications for improving both direct and indirect policy responses in challenging, dynamic systemic risk contexts, such as the COVID-19 pandemic.

Share this