These are hair-raising times. COVID-19 is a health crisis beyond what we've seen in decades. The fallout from the disease and from the countermeasures that are being taken touch every sector of the economy. Markets have responded with a significant fall in prices and a significant increase in volatility.
It is not a time to be relying on traditional models of market behavior. As John Maynard Keynes famously observed: "Economists set themselves too easy, too useless a task, if in tempestuous seasons they can only tell us, that when the storm is long past, the ocean is flat again." There are some periods when market forces bring prices somewhat close to the calm waters of the efficient equilibrium of classical economics. In those periods, the traditional models may provide useful insights on some decisions despite being just an approximation of a messier reality. This, however, is not one of those periods.
When there's a coronavirus-sized shock to the system, we need to look elsewhere if we are to make sense of the questions that matter. We need to look to complexity science. The study of complex systems has advanced significantly in the past 20 or 30 years. Complex systems are all around us — our brains are complex systems, the weather is a complex system, cities are complex systems, the natural world is a really big complex system. Rather than drawing only on the models and equations of physics (as economics has done since the 1950s), complexity science uses a wider set of tools.
The behavior of a complex system is driven by the interactions between its constituents. A bee colony is not just a collection of individual bees; the colony's collective behavior — establishing new hives, feeding and protecting, creating the next generation — emerges from how every action influences others, cascading through self-reinforcing feedback loops. Similarly, markets and economies are not just collections of individual investors. Market behavior — bubbles, crashes and everything in between — emerges out of analogous feedback loops. You can't develop an understanding of the behavior of a complex system by breaking it down into its subcomponents; you need to look at the whole system.
Behind much of the recent progress in understanding of complex systems is the work of the Santa Fe Institute, a think tank based in the both physically and intellectually rarefied air of New Mexico. This is a truly interdisciplinary entity — a rarity in the highly segmented world of academia — enabling invaluable cross-pollination. At an SFI gathering, you really don't know if you're going to be sat next to a biologist, a programmer, an archaeologist, a financier or an artist. That is exactly how it should be, in a field that is about the study of the whole system.
COVID-19 is massively disruptive for markets. It's not just that the impact is large, it's also that the impact is widespread, touching just about every corner of the economy. Each effect creates its own knock-on effects; these knock-on effects interact and compound each other, making the final outcome impossible to predict with confidence. This disruptive force has been unleashed on an economy made all the more fragile by just-in-time supply chains, outsourcing, short-termism and excessive fine-tuning. The investor community has become more integrated and more global in the past couple of decades, creating homogeneity of thought and strategy. Conformity of investor behavior means that liquidity can vanish just when it's most needed.
COVID-19 will reshape the environment within which these effects are playing out — a classic feedback loop. Political and social norms are being redefined, work practices reconfigured, priorities rethought. Some industries are likely to be permanently reshaped. In time, new industries will emerge; as SFI contributor Douglas Erwin has observed: "Mass extinctions create new evolutionary opportunities."
Famously, in a complex system, the flap of a butterfly's wings in the Amazon — or indeed the mutation of a virus in central China — can lead to a hurricane in Texas. The hurricane is not the end of the story, however. There is no stable norm to which everything will one day revert. The study of complex systems is a study of journeys, not of destinations, because when a system is in perpetual disequilibrium, there is only a journey.
No single model can fully capture complex system behavior, but models can nonetheless provide useful insights. For example, Richard Bookstaber describes in his 2017 book "The End of Theory" how agent-based modeling can be used to gain a better understanding of financial crises. Agent-based modeling assumes that individuals both respond to and shape their immediate environment, hence creating a dynamic and evolving system. This technique has been effective in the study of systems as diverse as stock market behavior, traffic flow and the movement of a flock of birds.
Agent-based modeling seeks insights rather than definitive answers: in a complex system, there are no definitive answers. As Mr. Bookstaber puts it: "The point isn't to crank out and act on a number. It is to set up a model to see what light can be shed on a real–world problem, and to see if it can fit a larger, intuitive narrative about what is going on." In other words, models should serve as an aid to understanding and not a replacement for it. No model is going to tell you exactly how COVID-19 will play out for markets, but well-designed agent-based models can identify key dynamics.
Just as traditional models work better in some situations than others, agent-based modeling is not right for every question. In reality, multiple models are needed. Mr. Bookstaber is blunt on this point: "If you've got one model, you are a dinosaur." At present, the most important questions are about how the disruption will work its way through the system. Where are the pressure points? Where could breakdown occur? How can it be avoided?
Complexity science has much to teach us about how systems behave at times of disruption. For investors who choose to learn the lesson, COVID-19 will spur the development of new ways to look at risk and at market behavior.
