Fixed link
0
vote

The Crisis of Global Capitalism

George Soros
Little, Brown & Co., 199

posted on 30 June 2005

reviewed by Joseph L. McCauley

The ideas in this book are based on Soros’ experience as successful speculator and market observer, and are described further in “The Alchemy of Finance”. One basic idea is that financial markets are inherently unstable, and deregulation makes them even more unstable. Second, but related, there are social needs that cannot be met by ‘market solutions’. This is the opposite of what’s taught in standard economics texts, where the Pareto Optimum is presented as if it would describe real markets (empirical evidence indicates strongly that it does not). Recent empirical research shows that Soros is right: financial markets are neither stable nor show any tendency toward either dynamic or statistical equilibrium. According to Soros, imposing ‘market discipline’ means imposing market instability.





Soros presents his argument in the interest of preserving the global capitalist system, as opposed to the desire to destroy it. Rightwing pundits, blinded by free market ideology, are unable to read his words objectively. The central point of the book is that “market fundamentalism” (radical free market ideology, or free market extremism) is a greater threat to a free society today than is totalitarianism. The book was written in 1998, before Islamic Fundamentalism in the Mid-East, and Christian Fundamentalism in the US, made their aims fully apparent to the rest of us.





Pointing out that preferences are not stable, which is opposite the notion of time-invariant preferences taught in mathematized neo-classical economic ideology, Soros contrasts two extreme societies: ‘transactional society’, where everything is freely traded and all values are priced in dollars, and a society that believes in ‘fundamental values’, as in fundamentalist religion. His main point is ‘reflexivity’, that in social phenomena what we think about affects what happen, because we can make wishes come true by acting on them (a problem of self-reference), and because we can never foresee all the consequences of our actions (incomplete knowledge). This is not the case in natural science, where thinking about an orbit of a planet cannot affect that orbit (unless we would fire a large enough nuclear rocket to shift the planet’s orbit). Soros as trader is always sceptical because he knows that there is always a gap between reality and his present perception of reality, that no one can be in possession of the absolute truth. This awareness emphasizes the danger posed by ‘Fundamentalist religion. His main point is ‘reflexivity’, that in social phenomena what we think about affects what happen, because we can make wishes come true by acting on them (a problem of self-reference), and because we can never foresee all the consequences of our actions (incomplete knowledge). This is not the case in natural science, where thinking about an orbit of a planet cannot affect that orbit (unless we would fire a large enough nuclear rocket to shift the planet’s orbit). Soros as trader is always sceptical because he knows that there is always a gap between reality and his present perception of reality, that no one can be in possession of the absolute truth. This awareness emphasizes the danger posed by ‘Fundamentalism’ or ‘Fundamental Values’ of all brands, if one imagines that those values represent inviolable truth. Likewise, transactional society (money-based society) is unstable, chaotic, for lack of any values other than ‘price’. His idea of an ‘Open Society’ lies somewhere in the no man’s land between those two extremes, and is based on the recognition of human fallibility, on the fact that no one is ever in possession of the ultimate socio-economic-religious truth, nor can anyone ever be.





Soros was strongly influenced by Popper in his early years, and gives much weight to falsifiability, which he thinks is limited in the social sphere to finding out eventually that one or more of one’s perceptions is wrong, and why. But he is optimistic tha, via scepticism, we can iterate toward social truth, even if we cannot ultimately reach it. E.g., a financial hypothesis doesn’t have to be true to be profitable, but if one doesn’t realize that the hypothesis is flawed and get out of the market fast enough, then one will get burned. Soros calls such gaps between perception and reality ‘fertile fallacies’. Osborne-Black-Scholes was a fertile fallacy until 1987, after which point options traders said it was wrong. The market model that is correct in our era is another fertile fallacy. The assumption from 1990-2000 that the market would only go up was a fertile fallacy, a much more profitable one than a mathematical market model!





Soros says that the events in which we participate do not constitute some sort of independent criterion (as in hard science) by which the truth or falsehood of our thoughts can be judged. By this, he means that there is always a gap between reality and how we perceive reality. However, he is not entirely right that Popperian falsification is beyond the pale: if ‘reflexivity’ is correct, and if we could mathematize it as an agent-based trading model, then we could test the idea on historic finance statistics. Why? Because if reflexivity is correct) then it must have produced those statistics! This mightn’t bring us a whit closer to predicting the future, and surely would not, but at least Soros’ idea can perhaps be tested, it may be falsifiable after all.





Now for the many mistakes in the text. Neither chaos theory nor evolutionary models have anything to do with exploring phenomena that cannot be determined by timeless laws (pg. 12). Newtonian mechanics is full of chaotic examples, and ‘evolutionary models’ (‘complex adaptable systems’) have so far explained absolutely nothing about biology. Logical positivism and Deconstruction/Postmodernism are certainly not the only worldviews available! Wigner (1967) has explained implicitly the difference between the hard and social sciences in his essays “Symmetries and Reflections”. A model needn’t be a timelessly-valid law of motion (like Newton’s Laws) to be falsifiable (pg.30), it need (like the Gunaratne-McCauley finance market model) only be empirically correct over a large enough time interval. Our model is correct over the last fifteen years, e.g., and is falsifiable. The following errors occur on pg. 36: pendula are not ‘ergodic’, movement toward ‘equilibrium’ occurs in no known real market, Osborne (1973) gave the correct definition of observable demand-supply curves, they are noninvertible step-functions where, e.g., x=D(p,t) exists but p=f(x,t) never exists. Stock market limit orders provide the canonical example. Pg. 39: equilibrium is not a special case of reflexivity, equilibrium does not exist in markets, which are both far from equilibrium and complex. Sunspot experiments did not establish General Relativity. On pg. 42 Soros disparages Black-Scholes, which wais a good zeroth order model and was falsifiable, and gives credence the economists’ impotent notion of ‘equilibrium’, which is not only falsified by real markets but was never true of any known market. Soros is not close to finance theorists (no disadvantage in making money in the markt) but is apparently close to two or more NYU economists, and is still too much influenced by the ‘equilibrium’ ideas taught in standard economics textbooks. “Dynamic disequilibrium” should read “disequilibrium”, there being no such thing as ‘static disequilibrium’ (another term that he uses elsewhere). A good dose of Newtonian physics, or at least a basic course in dynamical systems earlier at the London School of Economics might have prevented such misconceptions. “Value” is discussed on pg. 43, but we have shown elsewhere that there is no time-invariant definition of the (fundamental) ‘value’ of an asset, there is only the most probable or ‘consensus’ value attributed by most traders at a given time. Consensus value fluctuates much more wildly than the market statistics that are distributed about it. On pg. 49 Soros gives a wrong definition of equilibrium, analogous to one that Black apparently had in mind in his paper “Noise”. Pg. 55, market statistics have nothing to do with fractals, but are described to zeroth order by nonGaussian (nonlognornal, non everything you’ve read if you haven’t yet read Gunaratne and McCauley) ‘Brownian motion’ still with Hurst exponent H=1/2, in agreement with the efficient market hypothesis, which simply means that the market is hard to beat (shows no systematic patterns, to zeroth order). Pg. 59, no known market phenomena are susceptible to ‘equilibrium analysis’. Pg. 62, ‘static disequilibrium’ should be replaced by ‘closed system with small fluctuations’, maybe equilibrium, maybe not. Indeed, the only known way to reach statistical equilibrium in an otherwise free market is to impose upper and lower price controls (see “Dynamics of Markets” by JMC). By “dynamic disequilibrium’ Soros sometimes means far from equilibrium markets with volatility and fat tails (normal liquid markets), and sometimes he means market crashes. Market bubbles fall into the former category of a normal liquid market. On pg.66, another wrong definition of ‘equilibrium’ is given. There, Soros only means the gap between appearances and (the unknown) reality of the market that will make itself known in the future. Pg. 67, the claim about near and far from equilibrium for water is physically completely wrong. Pg. 70, there is no evidence that ‘open society represents near equilibrium conditions’, we know nothing of the sort. On pg. 90 is discussed valuing everything via money, and Spengler’s “Untergang des Abendlandes” could (and should) have been acknowledged.





In general, I’m in strong agreement with this book and recommend it very highly, not only to laymen but especially to economists and econophyicists, many (if not most) of whom have bad misconceptions about real markets. In particular, there is the myth of stationary markets, or asymptotic ‘market equilibrium’ in the face of nothing but market instability, as shown by hard empirical data. I would like to have been invited to correct the wrong definitions and misleading terminology in the book, but apparently that task was given to Economics Professor Roman Frydman at NYU. But now I have his promise that I get the job next time around.





References



1. G. Soros, The Alchemy of Finance, Wiley, NY, 1994.




2. E.P. Wigner, Symmetries and Reflections, Univ. Indiana, Bloomington, 1967.




3. J.L. McCauley, Dynamics of Markets, Cambridge, Cambridge, 2004.