Resilience and systemic risk

Standard risk management tools generally hold the implicit assumption that risks are normalized. Indeed the usual representation of risks on a matrix showing probability on one axis against impact on the other requires the risk to be quantified in the first place. Quantification is possible for many risks, but not for all.

Systemic risks are different – or more precisely risks in complex systems are different. Such systems are far from equilibrium and exhibit fat tail risks – i.e. rare high-impact events. Taleb eloquently popularized this familiar concept from statistics as black swans (although obfuscating the fact that swan colour is statistically predictable!). An important drawback of fat tail risks is that their probability is not defined – it cannot be assigned a numerical value. As a consequence, they need to be dealt with in a different way from the normal risk management process.

These risks may not occur often, but they are far from rare. In fact they are part of all familiar complex systems such as the economy, the financial markets, our bodies, healthcare, the climate, ecosystems, politics – in short they are a common feature of our day-to-day world.

Fortunately complex systems have a property that helps deal with these risks: resilience. We have developed a set of tools that assist management teams assess the resilience of a system. The practical result is that the risk conversation should be held in two distinct parts: The first looking at the standard tools for normalized risk, and the second looking at the resilience of the system with its dedicated systemic risk tools.


The Economist has written: “Farewell, left versus right. The contest that matters now is open against closed”.

In other places I have written extensively about how the traditional left/right divide has ended up conflating means and ends. The great human debate between more solidarity and more individual freedom has led to a tug of war between more state and more market. A quick scan around the world shows that this is a flawed equation: state and market are tools to an end, not the end itself. Market and state are better framed as interdependent: how they interwoven, defines what outcomes they deliver. Both the pure state and market frames are largely reductionist perspectives. Seeing at them as co-dependent leads a more systemic view.

Now the main political polarity is shifting. Coined as morphing from left/right to closed/open or from solidarity/freedom to protectionist/globalist. The US elections and Brexit are obvious recent illustrations. This shift, however, continues to conflate means and ends, albeit in a much more interesting way.

Closed and open describe types of systems. Complex systems are open, resilient and self-organizing. But subject to uncertainty, unforeseen effects and require more adaptive governance. Closed systems on the other hand are indeed much more controllable through top-down governance, but they often extract a price in other dimensions. Both systems can deliver various levels of wealth, inequality and happiness. The choice is not neutral, as it does appear open systems are better at delivering the goods in the long term. But there are lots of choices within each system, depending on the outcomes one prioritises.

So just as the market/state dichotomy confused means and ends, so does open/closed. The new framing does have the added benefit of inviting a debate on the system dynamics that might deliver desired outcomes.

Progress of a kind.

Published October 16th, 2016

Applying a complexity lens to the Dutch energy transition (or lack thereof)

The Kingdom of Orange is actually not very green. The Netherlands has become an EU laggard in environmental performance. While it still benefits from its earlier reputation as an environmental innovator, it has – in fact – fallen very far behind.

An energy transition is primarily about transitioning the energy system, right? Perhaps not. Together with collaborators from the Dutch Scientific Council for Government Policy (WRR) we published a report earlier this year, taking an alternative view.

A Dutch energy transition ought to be simple: the country is amongst the wealthiest in the EU, it can borrow at negative interest rates and has efficient public governance. While studies abound on how to shift its energy system, we took a different approach. We assumed that progress was being tied down by the deep interconnections of the energy system with other societal systems. You can’t change the one, without changing the others, by addressing their sticky path dependencies. In fact the logistics, industrial and horticulture backbones of the country are very fossil-fuel intensive.

Through a number of interactive workshops with experts from multiple sectors, we listed possible initiatives that could plausibly shift the system, shift social norms and accelerate decarbonisation. The goal was not necessarily to find the silver bullet, but rather to introduce a language and a frame for complex systems to assist policy makers widen their horizon – and identify new relevant solutions.
Wie is de wolf? Cover

Complexity in schools

In the Huffington Post, noted education blogger Cathy Rubin publishes an interview that we did a few months ago. In it i underscore the importance for a next generation of kids to grow up being complexity-litterate.

In fact working together with the International Baccalaureate Organisation, this appears quite doable. The feedback from many teachers we engaged with is that kids are natural systems thinkers – and they would welcome a curriculum that builds on that natural capacity, rather than try to replace it with a reductionist world-view. We are currently developing a complexity curriculum, in partnership with several IB schools.

Here is a link to the interview: Huffington Post

Methane is different @ COP21

Methane accounts for a third of the radiative forcing that warms the world. Yet for mitigation purposes under COP21 its impact is simply translated into CO2. That is likely not the best approach.

The form of governance needs to fit the dynamics of the underlying system.

Methane is emitted from agriculture, waste and the oil&gas industry, in roughly similar measure. See my report “The other knob:Tackling methane emissions presented at COP21.

With COP21 the theory of change for climate mitigation has shifted from a binding top-down approach in Kyoto, to a more dynamic bottom-up framework. Elinor Ostrom (Ostrom 2009) has argued that UNFCC top-down approach was not appropriate for the governance of the carbon commons. In a lucid paper for the World Bank, she argues for a more polycentric approach. In Complexity and the Art of Public Policy (Colander and Kupers 2014) the case is made for why certain systemic problems are best addressed with ‘ecostructure policies’, a combination of diverse bottom-up and strategic top-down action. The shift in approach at COP21 is consistent with these views.

This holds for CO2 and methane from agriculture. However methane emissions from the oil&gas industry are highly concentrated in relatively few players and largely disconnected from other systems – so they would paradoxically better be managed in the earlier top-down manner. One option presented may be pricing methane emissions independently from CO2.