Some thoughts on global risk

I’ve just spent a month looking at risk interconnections and have started to have some scary thoughts as a result. There is no doubt that global risks (at least as the Forum defines them) are highly interconnected. And as globalisation continues those interconnections are increasing and becoming stronger. As a simple example, as countries seek to manage their water scarcity issues by importing food on global markets, water risks and food risks go from being local concerns to impacting, and being impacted by, global supply chains and food price volatility.

What does it mean that we live in a world of increasing interconnections, particularly among global risks? Obviously it means that people are gaining in some way from these interconnections – they are able to trade more and access new markets, or lower transaction costs, or even offset specific strategic or operational risks in the short term. It also indicates that uncertainty is rising along with complexity as we are less able to identify the type or origins of potential risks, let alone the probability of their occurring. But most significantly it implies that we (individuals, organisations and states) are much more likely (perhaps alarmingly so) to be impacted by global risks.

Why? Clearly interconnections means contagion between risks and those affected is increasing: linked risks increase contagion from one event to another, while linked stakeholders and geographies spread it. But as well as the threat of risks propagating through a system more quickly and to more people, there is the issue that risks can be seen to arise endogenously from the system itself given a certain configuration of elements. If we assume that certain “patterns” of factors result in the emergence of a global risk, as linkages rise the total number of possible patterns that could result in a global risk multiplies rapidly.

Cleary and Malleret discuss this in their 2007 book “Global Risk” in the context of rising uncertainty. They point out that as the number of linked elements rises arithmetically, the number of interconnections in the system rises geometrically. The number of possible patterns therefore increases exponentially. The argue that the number of possible patterns in the system is 2^L where L is the number of links in the system.

Now in real life the networks that are relevant to global risks are not fully connected as this equation assumes, nor are they randomly connected with some evenly distributed average of connections. Instead, from what I remember from reading Paul Ormerod and Duncan Watts etc, they display non-linear, power-law properties stemming from the fact that some nodes are vastly more tightly connected than others, and connections can be of different strengths and in different directions.

But these points do not necessarily make increasing interconnections of global risks any less worrisome. While an additional connection at the margin (a weak node, weakly interconnected to the system) might only slightly increase risk levels, any connection that, either randomly or deliberately, has a non-trivial impact on a more important node could have serious implications for you or me. Any increase in interconnection means the possibility of an additional trigger, or new pattern that creates the necessary environment for a global risk to emerge. The interconnections then mean that this event will also have a wider impact.

This line of thought reminds me of Clay Shirky’s arguments about the negative returns from increasing complexity. He argues that at some point societies get to the point where increasing complexity starts to destroy value rather than add to it. But he adds that it is almost impossible for societies to *reduce* complexity. Hence, collapse is the result of “ratcheting” complexity. Which perhaps makes the WTO’s ratchet mechanism for increasing global trade dangerous for reasons other than many anti-globalisation protesters think.

But the bottom line is this: it seems reasonable to expect to be hit more and more often by volatility in the economy, financial and commodity markets, social structure, political scene and even weather, due to increasing rates of global risk events that are driven by the fact that the world is getting more interconnected. I’m going to go and check the maths and literature to see if there is an theoretical estimate for how quickly global risk exposure might be rising (just for interest’s sake), but it might nevertheless be worthwhile considering what you can do to be personally more resilient to system shocks. And it also might be time for us to think harder about what improvements to the system as a whole might increase overall resilience without foregoing the advantages of global trade and communications. And of course to hope that we aren’t entering a world where increasing complexity is dooming us to collapse!