Customer Reviews

The Self-Organizing Economy

5 star:		(1)
4 star:		(1)
3 star:		(0)
2 star:		(1)
1 star:		(1)

 

 

This is a delightful little read, bringing bits of complex-systems analysis to economics. It requires little in the way of mathematics. Because it's Paul Krugman, it's effortlessly charming, self-deprecating, and eloquent.

Those who are only familiar with Krugman's excellent New York Times op-ed pieces may not know that he was (and still is?) a world-renowed economist whose work centered on "economic geography" -- i.e., understanding why it is that economic activity on so many scales is localized rather than dispersed. Why do cities exist? Why do they exist where they do? Why is economic activity so concentrated in the West? Why does Africa have such a hard time becoming part of the "developed world"? Clearly there are connections between economic geography and international development. A complete theory of economic geography would explain not only why activity concentrates where it does, but what policymakers can do to bring activity to inactive areas.

The Self-Organizing Economy starts with some simple models of geographic concentration. Suppose that there's a certain benefit to moving your business near an economic center like a city; the benefit may be something as clear-cut as reduced transportation costs. But being further away from a city has its benefits as well, in decreased rent. Tune your parameters accordingly, to qualitatively match observed behavior. Start with a hypothetical land area where businesses are located at random, then use a computer to simulate how the distribution of businesses will change over time. You'll find -- Krugman did -- that concentration is inevitable over a wide range of parameters.

He starts with the segregation model from Thomas Schelling's Micromotives and Macrobehavior. You don't even need to assume that people are explicitly racist, in the sense that they'd prefer a neighborhood where everyone looks like them. You just need to make the weaker assumption that people would prefer not to be the lone white person in a black neighborhood. Under that assumption, a perfectly integrated neighborhood is not a stable equilibrium as the process evolves in time: a slight perturbation (which in this case would mean something like "a few more black people move in") throws the system quickly away from that equilibrium and on to another one.

He moves on to macroeconomics, namely trying to understand why business cycles happen at all. Why doesn't the economy grow consistently with population? Why does it instead go through periods of boom and bust? There's an interesting hypothesis in here that inventories always lie near an unstable equilibrium, and that an external (in the jargon: exogenous) shock is sufficient to knock us away from that equilibrium.

Connected with which, there's an even more interesting story (Krugman doesn't bother calling it a "model") to make explicit exactly what "globalization" means. In terms of actual economic value, Krugman has persuasively argued in "Pop Internationalism" that we are not a "global economy," and in any case that the economy was more global before World War I. He hits the same theme in The Self-Organizing Economy, just long enough to tie globalization in with ideas from complex systems. Other nations' economies are also stuck near an unstable equilibrium, for the same reason that ours is. A shock that knocks us into a recession knocks them into one, too; our business cycle is "phase-locked" with theirs.

The first 2/3 of the book is all prose: the best intuitive presentation imaginable for the topics Krugman covers. Integrals are included here and there in the last third, but they're not vital to understanding the material.

Paul Krugman stands out again from the poseurs, charlatans and know-it-alls who think they are the first people to have really understood economics. This book has the quality of an occasional piece, of reflections on the subject rather than an in-depth research monograph, but still the insights just keep on coming.

Maybe reviewers who tack on economics and finance at the end of their list of research "specialties" -- "nonlinear dynamics (especially classical mechanics), statistical mechanics, cosmology, superfluids, hydrodynamics and turbulence, porous media, economics and finance" -- understood all the applications of non-linear dynamics and complexity theory to economies, but for the rest of us this little book is a delight.

Self-Organising is a well-developped concept in physics and Krugman tries to apply that concept to economics. Despite its interesting title, knowledgable readers will find that there is no real insight offered, other than some common sense discussions. Krugman in the book claimed that before "the visionaries of Santa Fe" economists have tried long ago using "nonlinearity" in modelling economics. Of course nonlinear functions have been used since day one in economics, the question is how to use it CORRECTLY. Krugman's book is just one more example how nonlinearity can be misused.

I still recommend reading this book, at least to understand why the current economics is so boring. Best read together with the two volumes of Santa Fe proceedings (Anderson and Arrow 1987; and B. Arthur et al 1996).

A recent article is revealing on why economics is becoming more and more dismal despite the attempts of putting a new face (like this book): John Cassidy "The decline of economics", The New Yorker Dec.2 1996.

'Self-Organized' is a widely-used but not precisely defined term. Krugman was inspired by the idea of self-organized criticality made popular by a certain school of physics, but there is nothing original in his book. He has merely lifted notions from another field that he did not invent and, far worse, doesn't understand. For something much better, and which is at least explanatory and seems to make some sense, try his book 'The Return of Depression Economics'. 'The Self-Organizing Economy' is best tossed in the trash can. I confess that the title of my review is not original, but I refuse to divulge the author in order to protect his privacy.

For critical analyses by economists of what is wrong with the basic theory believed by Krugman (and the entire MIT and Chicago schools, the IMF, The World Bank, all so-called neo-classical economics), see Paul Ormerod's 'The Death of Economics'. Like all neo-classical economists, is a utility maximizer, a believer in stable equilibria, but real data (which are always far from equilibrium) show utility maximization to make terribly wrong predictions, not even close to the mark, completely useless except as a way to mislead lawyers and politicians who understand too little to question the results. Also, do not miss reading Mirowski's Machine Dreams-fantastic book!