Beyond Mechanical Markets: Asset Price Swings, Risk, and the Role of the State
Roman Frydman
Four stars for the concepts, three stars for the book. I first learned about Roman Frydman from watching a video of his presentation at the Berlin conference of INET, a somewhat unorthodox economics association. The research program embodied in this book strikes me as extremely valuable and promising for economics. Basically, almost all modern economic writing on financial markets begins from the assumption that asset prices and/or fundamental factors evolve, even if unpredictably, according to known probability distributions (the “ergodic” assumption). Second, the economist assumes that all agents being modeled must know the model being posited as the “true” model, and thus make their forecasts consistently with it (the “rational expectations hypothesis”). Frydman’s argument follows Keynes and other early moderns in seeing “non-routine change” (i.e., that which cannot be modeled by ergodic assumptions) as the fundamental fact of capitalism and of financial markets.
Of course, throwing out the ergodic assumption and the REH is a huge setback for the economist-as-modeler, since it’s very unclear how it would be possible to say anything useful about inherently unpredictable change. But I very much appreciate the effort to start moving in that direction–to stop looking for the keys under the streetlamp, as it were. This book is the non-technical version of the material, so I can’t completely judge the modeling approach, but it strikes me as both reasonable and useful.
I really appreciated the analogy drawn between ergodic modeling and the Soviet attempt at a planned economy. It is certainly a provocative one, but I think it does highlight useful parallels and makes useful points about the distribution of knowledge.
One thing I didn’t like as much was the section on possible policy responses to asset-price swings. It seemed less carefully thought-through than the other sections of the book. Perhaps it is de rigueur for a popular-consumption book to have a “policy angle,” and perhaps the policies discussed really would be useful, but there didn’t seem to be enough space to develop them as really compelling arguments.