The similar dynamics of tsunamis and earthquakes and ruptures in manmade
system such as economies and electric power grids may have profound
implications for future planners and policy makers."Power Curves: What Natural and Economic Disasters Have in Common,"
a McKinsey Quarterlyarticle by Michele Zanini, describes how
complexity theory explored in the natural sciences is increasingly being
used in economics. In fact, she writes, Didier Sornette,
an earthquake authority and visiting geophysics professor at University of California at Los Angeles, is heading the Financial Crisis Observatory in
Zurich, which uses mathematical models based on complexity science and
physics to gain insights into economic crises. Power laws, she writes,
seem applicable to earthquakes,
forest fires, blackouts and banking crises. For example, Ms. Zanini
says, plotting the frequency of banking crises around the world from
1970 to 2007 and the magnitude of their loses in each country, shows a
power law curve: a short "head" of nearly 70 crisis with relatively
small losses and a very long "tail" with a very small number of massive
Much has been written about the concept of power laws. Nassim Nicholas Taleb's engaging and insightful 2007 book, The Black Swan: The Impact of the Highly Improbable
has been a best-seller. Sornette proposes "dragon kings"-
outlier events that he says are even wilder than Taleb's black swans.
what does this mean for the possibility of predicting? This field of
research is still young, but Ms. Zanini writes that what's known so far
has potential implications. First, she says, "make the system the unit
of analysis", meaning that the behavior of any agent has to be studied
in the context of the behavior and performance of the system in which
it is embedded. She also suggests paying attention to history, noting
early warnings, learning from other complex systems, and building
flexible business models.
Research scientist and author Duncan Watts
has some similar observations in his essay "Too Complex to Exist,"
urging analysts to study "systemic risk." He cites a stunning example.
In 1996, a single power line in Oregon brushed against a tree and
shorted out, triggering a massive cascade of power outages that left 10
million people in the Western US without electricity. No one figured
out exactly how it happened ,or how to prevent it. And a similar
blackout cascade happened again in the Northeast in 2003. He writes that
the financial system, much like the power grid, "is a series of
complex, interlocking contingencies." Once a cascade starts in any
complex system, he suggests, risk assessment is very hard because
conditions influencing any part of the interconnected system can change
suddenly and dramatically. "...The number of contingencies that a
systemic risk model must anticipate grows exponentially with the
connectivity of the system," he writes.
Watts says there has
been a trend toward building ever larger and more connected networks,
and he suggests that "too big to fail" may really mean "too big to
exist." One possibility, he writes, could be a regular review of the
largest and most connected firms in each industry by regulators who
could get answers to an essential question: Would the sudden failure of
this company generate intolerable impact on the wider economy?
Government intervention in the economy already exists, he observes, and
the real question is what kind of intervention works best: "preventive
management or after-the-fact rescue."