Using Network Theory to Understand Crises

(Click for larger image. The U.S. economy as a network, divided by sectors: Technology (blue), oil (dark gray), other basic materials (light gray), finance linked to real estate (dark green), other finance (light green))

Brandon Keim at Wired covered some new work in which researchers are examining the U.S. economy as a network in order to better understand the 2008 financial crisis and why it had such a devastating effect on the real economy.

From this analysis came two striking figures. The first is a map [above] of links between companies in five key economic sectors: technology, oil, other basic materials, finance linked to real estate and other finance. As of 2003, the sectors are relatively distinct, with real estate isolated. By 2008, they’re a tightly linked jumble, with finance at the center. ...

Other research on network dynamics has shown that interdependence can promote stability, but eventually reaches a point of reversing returns (see “Networked Networks Are Prone to Epic Failure“).

Much more at the links, so please click through. To me, there are two ways of looking at this. The first is the conclusion reached by Keim, that interdependence on its own can be stabilizing, until it reaches a critical mass, at which point increased interdependence destabilizes the system. Interdependence obviously went up throughout the 2000s. But another way to look at it is to examine the pattern of interdependence, rather than the occurrence of interdependence.

It is clear that the financial sector became much more central to the economy, so the economy as a whole became much more susceptible to trouble in the financial sector. In this way, the U.S. economy appears to display a feature of non-random, hierarchical networks, which is that they are robust to shocks in peripheral parts of the network, but fragile to shocks at the center. In other words, if a shock had hit the peripheral oil sector (as happened, in fact, in the middle part of the decade), the increased interlinkages with finance would make the economy more resilient. But once a shock hit finance, the central sector, everything else was prone to collapse as well.

All of us at IPE@UNC (plus Andy Pennock, who's on the job market this year, as hiring departments should note) are working on a project in a somewhat-similar vein, but in an international context. I'm sure we'll be blogging much more along these lines in the coming months. In any case, it's fascinating stuff and is being used more and more to try to understand the complex interdependencies in the global economy. The political implications of this work is also important, though I haven't much mentioned them here.

(ht: Josh Miller)