Welcome to European Tribune. It's gone a bit quiet around here these days, but it's still going.

More (Macro)Economics by Physicists

by marco Sun Nov 27th, 2005 at 10:59:52 PM EST

Reading Migeru's 11/13 diary on "Division of labour and comparative advantage: what linearity hides in economics", I was reminded of another paper on macroeconomics, this one written by Jean-Philippe Bouchaud and Marc Mézard, "condensed matter" physicists who attempt to explain Pareto's principle "using both a mapping to the random `directed polymer' problem, as well as numerical simulations".

I was interested in what seemed to be the striking implications of the paper, primarily that "the basic inequality in wealth distribution seen in most societies may have little to do with differences in the backgrounds and talents of their citizens. Rather, the disparity appears to be something akin to a law of economic life that emerges naturally as an organizational feature of a network."

Also, that

  • a minority of people in any society will inevitably wind up with a disproportionately large amount of th wealth ("Chop off the heads of the rich, and a new rich will soon take their place");

  • increasing the "temperature" of an economy (i.e. how easy it is to trade in that economy) enables "wealth to flow easily from one person to another, tending to spread money more evenly"

  • increasing the "E" exponent in Pareto's law, which corresponds to a larger number of people in the society possessing more shares of the wealth, can be achieved though higher taxes that are redistributed "evenly", as well as by increasing the amount of money flowing throughthe system and how often it changes hands.

As a very very lay layperson both in economic as well as in physics, I am in no position to judge the merits, or even the novelty, of the Bouchaud and Mézard's work.  I did notice, however, that after an initial spate of interest in the popular science press, it all sort of fizzed out.  So naturally I wondered if perhaps the emperor had no clothes on after all.

If anyone is familiar with this paper or this work, I would be grateful to hear your opinions on them. (I originally read about their 2000 paper, "Wealth condensation in a simple model of economy", in an article titled "That's the way the money goes" by Mark Buchanan in New Scientist and found another article titled "Wealth Distribution and the Role of Networks" also by Buchanan in HBS Working Knowledge (unfortunately, the maths in the original paper are way above my head, as was much of the more mathematical content in Migeru's diary... so sad, i know. One page that contains several useful links is Peter Kaminski's entry on Wealth Distribution.)


Display:
I agree with the first two points, but not (wholly) with the third.  Redistribution via taxation -- let's just lump this under "fiscal policy" -- can help achieve greater equality, so long as you do not kill the incentive to work and achieve.  Risk-taking is a critical component to economic success.  Big risks, successful when taken, should be rewarded with big payoffs.  ("How big?," is the question to ask.)

Increasing the money supply and the number of transactions will, in the long run, lead only to rising prices if it is not met with greater aggregate wealth.  In the short run, it can help the the "little guy" -- e.g., the debtor -- but only in the short run.

Be nice to America. Or we'll bring democracy to your country.

by Drew J Jones (pedobear@pennstatefootball.com) on Mon Nov 28th, 2005 at 12:34:18 AM EST
It's a balancing act.
by Colman (colman at eurotrib.com) on Mon Nov 28th, 2005 at 06:17:54 AM EST
[ Parent ]
Right.

Be nice to America. Or we'll bring democracy to your country.
by Drew J Jones (pedobear@pennstatefootball.com) on Mon Nov 28th, 2005 at 01:09:07 PM EST
[ Parent ]
Can we learn something about wealth condensation in Norwegian lattitudes?

I guess that economic crises are good for wealth condensation, but FDR-type "New Deals" not so good ;-)

by das monde on Mon Nov 28th, 2005 at 01:36:24 AM EST
I was planning on writing a diary on self-organization and economics. I'll read their paper and build the diary around it.

A society committed to the notion that government is always bad will have bad government. And it doesn't have to be that way. — Paul Krugman
by Migeru (migeru at eurotrib dot com) on Mon Nov 28th, 2005 at 05:12:37 AM EST
I've had finding this paper on my to-do list since I read the NS article about it.
by Colman (colman at eurotrib.com) on Mon Nov 28th, 2005 at 06:18:51 AM EST
[ Parent ]
way above my head, as was much of the more mathematical content in Migeru's diary
The big secret about mathematics is that what makes it hard is not the equations.

A society committed to the notion that government is always bad will have bad government. And it doesn't have to be that way. — Paul Krugman
by Migeru (migeru at eurotrib dot com) on Mon Nov 28th, 2005 at 05:46:21 AM EST
The conclusions of the paper are not surprising: they are exactly what you would expect from general principles of modern condensed matter physics established in the 1970's.

What I wrote in my earlier diary is analogous to the kind of qualitative thermodynamic models physicists made around 1900 (I mentioned specifically the Van der Waals equation of state for fluids). These are purely qualitative, macroscopic models.

In the mid-20th century a lot of microscopic models were studied using a technique called mean field theory. This was a half-way house in which individual particles were considered, but each of them was only the average effect of all the others. These models were more successful qualitatively but quantitatively were not borne out by experiment near the so-called critical points (there is a very inadequate article over at Wikipedia --- I'm going to have to go and rewrite it).

There were two empirical facts that made critical points remarkable:

  • the relationships between measurable quantities could be expressed as power laws near the critical point
  • very different physical systems exhibited the same power laws (with the same accurately measured exponents)

These empirical facts led to the discovery of two concepts which have come to dominate theoretical physics for the last 35 years (and were worth a few Nobel Prizes):
  • scale invariance: the power laws observed in the vicinity of a critical point indicate that, at the critical point, the system under consideration becomes perfecly scale-invariant. [This is a subtle mathematical/geometrical insight that I can't do justice to here]
  • universality: the fact that the same numerical exponents were observed across very different physical systems led to the discovery of universality classes. It turns out that, depending on the dimensionality of what is called the order parameter and the dimensionality of the space on which the system lives, thermodynamical systems fall into specific universality classes with the same qualitative behaviour. In other words, close to the critical point the behaviour of a macroscopic material is independent of its microscopic details, not only qualitatively but also quantitatively.
I don't know whether you have to be a physicist to appreciate the import of these two insights, but they are truly mind-blowing.

Then in the 1980's it was realized that, when a system is far from equilibrium, it tends to arrange itself into a "critical state", where "critical" here just means that various quantities of interest follow power laws as in the classical critical points. Unlike the traditional critical points of equilibrium thermodynamics, which were highly unstable and delicate, these dynamical critical points were stable. Because the power laws indicate that there is structure at all scales, from the smallest to the largest, these critical points appear highly organized, and because of the stability they arise spontaneously in the right conditions. This led to the idea that systems out of equilibrium self-organize into a critical state. Because of universality, the microscopic details of the system should be largely irrelevant.

And what does this have to do with social sciences? Well, Pareto's law of income, or Zipf's law, or the distribution of the size of towns in a country..., all follow power laws. It is very easy for a physicist to jump to the conclusion thet self-organized criticality is at work. What is hard is to exhibit it through a specific quantitative model. What makes things a little easier is universality: to a large extent the microscopic details of the model don't matter.

The point to take home from all of this is that there is a very different paradigm in Physics right now from the mechanistic world of the 19th century that economists sometimes try to imitate. 21st century physics is hosptable to life where 19th century physics was bleak and lifeless. This new paradigm accommodates an evolving universe in which organization arises spontaneously and increases continuously, which is very different from the bleak heat-death of the universe view of the world we inherited from 19th-century physics. Also, the mechanistic clockwork universe of Newton and Laplace has been replaced by the  unpredictable (and organized) world of chaos and "strange" attractors (this world is still deterministic, but unpredictable and so not mechanistic).

Now I really need to go write a diary about this.

A society committed to the notion that government is always bad will have bad government. And it doesn't have to be that way. — Paul Krugman

by Migeru (migeru at eurotrib dot com) on Mon Nov 28th, 2005 at 05:07:49 PM EST
You certainly do need to write a diary about this!

Great post thanks.

by Metatone (metatone [a|t] gmail (dot) com) on Mon Nov 28th, 2005 at 05:27:25 PM EST
[ Parent ]
each of them was only the average effect of all the others
each of them saw only the average effect of all the others

A society committed to the notion that government is always bad will have bad government. And it doesn't have to be that way. — Paul Krugman
by Migeru (migeru at eurotrib dot com) on Mon Nov 28th, 2005 at 05:37:41 PM EST
[ Parent ]
Awesome post.  Thanks for that.  I share your excitement about these new, more dynamic paradigms that have been emerging in physics over the last few decades.  (Do you know Cosma Shalizi's blog, "The Three-Toed Sloth"?  He is a statistical physicist "building predictive models from data generated by stochastic processes, and applying those models to questions about neural information processing, self-organization in cellular automata, and so forth.  All of this is about using tools from probability, statistics, and machine learning to understand nonlinear dynamical systems and non-equilibrium statistical mechanics."  He tags each entry with pertinent category labels, such as Complexity, Networks, The Dismal Science, and The Collective Use and Evolution of Concepts, among many others outside of physics and science.)

I am very much looking forward to your diary, and hoping that you can provide some input as to how seriously one should take the Bouchard & Mézard paper in terms of what implications it may have for real world economics.

Point n'est besoin d'espérer pour entreprendre, ni de réussir pour persévérer. - Charles le Téméraire

by marco on Mon Nov 28th, 2005 at 06:18:32 PM EST
[ Parent ]
just wondering though, has anyone applied this to the stock market and broken the bank?  maybe some of the very successful hedge funds, or someone that is just cleaning up and keeping it to themselves?
by wchurchill on Tue Nov 29th, 2005 at 03:13:04 AM EST
This is not the point. If this theory is correct, it implies that the economy is essentially unpredictable.

A society committed to the notion that government is always bad will have bad government. And it doesn't have to be that way. — Paul Krugman
by Migeru (migeru at eurotrib dot com) on Tue Nov 29th, 2005 at 03:16:09 AM EST
[ Parent ]
theories in finance?
by wchurchill on Tue Nov 29th, 2005 at 03:22:54 AM EST
[ Parent ]


Display:
Go to: [ European Tribune Homepage : Top of page : Top of comments ]