04 November 2017
The last Roman emperor of the west, Romulus Augustulus, was deposed in the autumn of 476 AD. As the grapes ripened on the vines that year, the Roman Empire in western Europe, a system that had lasted for 500 years, collapsed. The Dark Ages began.
From the Roman Empire to international banking crises, human history shows that even vast networks, affecting billions of people, can be at risk of collapse. Their size does not protect them, and simple weaknesses, built into the system, can bring the whole structure down.
The risk of system collapse—known as systemic risk—came under close scrutiny from scientists after the bankruptcy of the Lehman Brothers in 2008 sent shockwaves through the financial world and connected banks began to fall like dominoes. Governments scrambled to reduce the impact, but the world slipped into a global recession.
In the wake of the crisis, IIASA researchers developed a method of measuring how risky each financial transaction is—in other words, how much it contributes to the chances and costs of system collapse. Equipped with this knowledge, it is possible to tax transactions according to the risk they pose to the system.
“If you tax systemically risky transactions, you give banks an incentive to avoid them, making the system more resilient,” says IIASA researcher Stefan Thurner, who spearheaded the approach. And testing with models indicates that it is extremely effective—when the tax is introduced, the system rewires itself into a form that is basically free of systemic risk.
Not only that, but it does so without reducing the amount traded. Taxes like the “Tobin tax”—proposed as a response to the 2008 crisis—reduce the number of transactions overall because they are charged indiscriminately on every transaction, rather than targeting those that increase risk.
“The fascinating thing about systemic risk tax is that it does not reduce the volume of transactions, but just re-shapes the network,” says Thurner. The idea has yet to be put into action in a real-world banking system, but it has already gained attention from the world’s central banks, who are interested in calculating the risks associated with their own transactions.
While the financial crisis focused attention on systemic risk, researchers can study numerous systems—from epidemics to food security to governance—using the same principles.
“When we study these systems we are looking at networks of nodes and links,” says Ulf Dieckmann, Evolution and Ecology Program director at IIASA. “In a financial system the nodes are banks and the links are transactions, in epidemiology the nodes are people and the links are cases of infection transmission, and in wildlife conservation the nodes might be isolated populations linked by dispersing animals. There is a huge diversity of applications for this work.”
Scientists have been warning of the risk of an epidemic sweeping across our highly connected world for many years. In an urban public transport network, the confined spaces and high numbers of people passing through—especially during the daily commute—mean that a virulent disease could make its way across a city in even a few hours.
The densely populated city of Tokyo is the perfect place to examine how a public transport system might spread a contagious illness and the countermeasures needed to protect inhabitants.
IIASA researcher Akira Sasaki and colleagues used the city as a case study to develop a “network centrality measure,” which assesses how important each network node—in this case, station—is in terms of disease spread.
Their findings make eye-opening reading, especially for public health officials seeking to understand how to allocate resources in the event of an outbreak. Targeting the largest station in the city with countermeasures is 1,000 times more effective in stopping disease spread than at the second largest station, they found, even though the number of people passing through is only around 1.5 times greater. The message is that when the disease strikes, decision makers need to pour resources heavily into the biggest station.
Our vast global trade networks can keep hunger at bay by allowing nations to import food when homegrown produce is scarce. But these same networks mean that a disruption to food production on one side of the planet may cause famine on the other.
In the 1980s, Alaskan boats were landing 240,000 tons of fish every year, which was exported to many distant countries. But in 1989, in one of the worst environmental disasters in history, the Exxon Valdez oil tanker foundered on Bligh Reef, pouring millions of tons of crude oil into the near-pristine ecosystem. The complete closure of the Alaskan fishery that followed led to fish shortages around the world.
Seafood plays an important role in food security, making up nearly 20% of animal protein consumption globally. And it is not just oil spills that can disrupt the seafood trade network. Overfishing caused the collapse of the northwest Atlantic cod fishery in 1992, when the population fell to just 1% of its original size. Disease, invasive species, and political turmoil can also reduce a prospering fishery to a shadow of its former self.
So when a shock hits the system who suffers the most? Jessica Gephart, a postdoc at the National Socio-Environmental Synthesis Center, USA, worked on this problem with IIASA colleagues during the institute’s Young Scientist Summer Program. “We wanted to understand which countries are the most vulnerable to these kinds of shocks. Our measure of vulnerability included how big a part seafood plays in their diets, and how much a shock to the seafood trade might reduce their imports,” says Gephart.
The results show that central and western Africa are at the greatest risk. “People in these countries rely on seafood as an important source of protein, and they are importers rather than exporters,” says IIASA researcher Åke Brännström, a coauthor on the paper.
On top of that, when seafood is scarce, prices go up, and people in rich countries can just pay more. Adding this dimension into the model showed that it puts the poorer countries in central and western Africa in an even worse position.
“To improve resilience, countries can try to diversify their diets and include other sources of protein. The sad thing is that this is difficult in developing countries, which often lack adaptive capacity. Another approach would be to try to increase the domestic supply of seafood, but care must be taken to avoid the environmental problems of overfishing and unsustainable aquaculture practices,” says Gephart.
Natural disasters provide perhaps the most powerful image of system collapse. A hurricane that flattens buildings, floods roads, and downs electricity grids has very effectively brought a system to its knees.
In South Asia this summer over 1,200 people were killed in flooding and landslides as a result of severe monsoon rains, with around 40 million people affected by the catastrophe. The media showed pictures of people wading through floodwaters clutching their few remaining possessions, or families huddled in the wreckage of their homes. But what follows a disaster like this often doesn’t make the headlines. The loss of prime agricultural land, damaged infrastructure, and disruption of children’s education make for starvation, poverty, and lost livelihoods long after the flood waters subside.
The solution is fundamentally the same as for the financial markets or global trade networks—resilience must be built into the system. “We can look at disaster risk, that’s about probability, but really a disaster like a flood or a storm is only the trigger,” says IIASA researcher Stefan Hochrainer-Stigler. “Systemic risk is different, it’s the risk to the whole system, and it stems from how the elements of a system are connected.”
To understand how a nation’s economy will cope when put under the stress of a natural disaster, IIASA researchers have been using groundbreaking agent-based modeling. “We are now able to model an entire national economy, including households, firms, and banks—the largest and most detailed model of its kind,” says Sebastian Poledna. “We can show how a disaster impact will spread across the economic system, and that gives us a starting point for understanding how to strengthen and protect that system.”
Ultimately, systemic risk comes from within. External forces—in the form of bankruptcy, disease, or flood—can expose the weaknesses in a system and even lead to its collapse, but preventing them is not necessarily the solution, and may not even be possible. In the end, the only way to protect a system might be to change it.
Text by Daisy Brickhill
Poledna S & Thurner S (2016). Elimination of systemic risk in financial networks by means of a systemic risk transaction tax. Quantitative Finance: 1-15. [pure.iiasa.ac.at/12653]
Leduc MV & Thurner S (2017). Incentivizing Resilience in Financial Networks. Journal of Economic Dynamics and Control 82: 44-66. [pure.iiasa.ac.at/14630]
Yashima K & Sasaki A (2016). Spotting epidemic keystones by R0 sensitivity analysis: high-risk stations in the Tokyo Metropolitan Area. PLOS One 11: e0162406. [pure.iiasa.ac.at/13919]
Gephart JA, Rovenskaya E, Dieckmann U, Pace ML, & Brännström Å (2016). Vulnerability to shocks in the global seafood trade network. Environmental Research Letters 11: e035008. [pure.iiasa.ac.at/12165]
Hochrainer-Stigler S & Poledna S (2016). Modelling Macroeconomic Effects of Natural Disaster Risk: A Large Scale Agent Based Modelling Approach Using Copulas. In: 7th International Conference on Integrated Disaster Risk Management Disasters and Development 2016 [pure.iiasa.ac.at/13875]
Last edited: 17 November 2017
International Institute for Applied Systems Analysis (IIASA)
Schlossplatz 1, A-2361 Laxenburg, Austria
Phone: (+43 2236) 807 0 Fax:(+43 2236) 71 313