Metastable States – the Meltdown Precursors

I’ve just read a recent column by JL, one of the editors from Taipan Daily. He states, in his column “There Will Be Blood in Europe”:

Stepping back a bit: What is so frightening right now, not just in Europe but China and America and Japan too, is the presence of fraud-fueled “Lehman 2.0” catalysts threatening to explode.

One could say that the 2008 financial crisis was the mother of all wake-up calls. But instead of actually waking up, the powers that be slammed the alarm clock, choked down a fistful of Ambien, and rolled back to sleep.

As a result, the world is going to get an even bigger wake-up call in the not-so-distant future.

This current Case Study is using the 2008-9 financial systems meltdown as the focal point. Starting today, I’m going to begin making the crucial parameter identifications that indicate when a metastable state will collapse, so that a “meltdown” occurs.

The most important thing to note right now is that – both in the model predictions AND in the real-world events that we’ve been observing – meltdowns happen fast. We can be in a metastable state that lasts so long, and is so extreme, that many people believe that the situation will last forever.

But it doesn’t.

This Case Study will show the underlying dynamics, and how these “meltdowns” are set up, and what happens when they collapse – all using the very simplest model possible from statistical thermodynamics.

In the last post, I characterized a state where very few institutions (“units” in the statistical thermodynamics model) were involved in risky (overly-leveraged) situations. This occurs when the free energy minimum occurs for a low value of x, where x is the decimal fraction of total units (institutions) involved in risky deals. It corresponds to Region A of the phase space diagram, shown two blogposts ago, and also previously.

With this posting, we move on to Region D, which is the one where metastabilities exist. That means that there are two free energy minima, throughout all points in Region D. (If you’ll refer to the phase diagram, you’ll see that Region D is the pink area in the middle; bordered by Region A at the top and Region G below, where both A and G are light blue.)

Note Added March 6, 2014: For a detailed walk-through of this journey through phase space, see the White Paper: Statistical Thermodynamics: Introduction to Phase Space and Metastable States, which discusses Phase Space and Metastable States for the Ising Spin Glass with Interactive Enthalpy.