on June 4, 2004
This book possesses two distinct features. It errs frequently and amateurishly. Perhaps the most important error in this book is in using the Arrow-Debreu model in discussing the debate over Socialism between Lange and Hayek. In the ADM scheme of things there is a complete set of markets where unrestricted competition among perfectly informed persons reigns. Naturally, such perfect conditions yield perfect results- on paper. Actual markets do not deliver these results so "the competitive paradigm is not robust" (p107). Of course, it makes no sense to think of the Lange-Hayek debate in terms of Arrow Debreu anyway. This debate over socialism took place mainly in the nineteen thirties. The ADM was published in 1954.
Socialism also suffers from incentive problems, so Stiglitz tells us that the Lange-Lerner model of market socialism does not depict reality either. In reality, imperfect information results in imperfections in both systems. So this book provides its readers with the "revelation" that real world institutions are not perfect. Another error is in Stiglitz near exclusive interest in mathematical equilibrium models that ignore dynamic issues in comparative economics. These errors lead him on the one hand to accept that Soviet type economies do not work well, but on the other hand that we should be wary of markets, and instead consider different forms of state intervention.
For those who have actually read and understood the writings of the old socialists, like Lange, Schumpeter, and Lerner, and their opponents Mises and Hayek this debate looks quite different. All of these economists recognized imperfections in both systems and wrote about them explicitly. Lange is well known for admitting to the danger of the bureaucratization of economic life under socialism. Hayek stressed the problems with information under both systems, and in fact spelled out the crucial informational imperfections in markets by 1937. Mises, Hayek, Lange, and Schumpeter all thought in dynamic terms that went far beyond the models with which Stiglitz likes to play. Stiglitz tries to reduce Schumpeter's process of creative destruction to a simple (and weak) game theoretic argument. I saw no mention of Mises in this book at all.
This book is lacking both in the breadth and the comprehension in its authors research. The main focus of this book is intead to play up the importance of its authors prior writings on imperfect information. All past accomplishments on both sides of this debate get recast into a form that makes the author of this book the only one who has made a worthwhile contribution, when in reality it is the contributions of intellectual giants like Hayek, Mises, Lerner and Schumpeter that rise above all else. Modest scholars recognize the shoulders that they stand upon from the past. The author of this book shows little such modesty, and even less in terms of important and original advances.
on January 23, 2002
Stiglitz shows how much things can change, when you drop assumtions like costless information, zero transaction costs etc. According to the general equilibrium theories he crizices, a central planner could archive an outcome that is at least as efficient or better than the market, by imitating perfect competition (Lange-Lerner-Taylor Theorem). Stiglitz shows that by dropping unrealistic assumptions both real markets and market socialism aren't that efficient as the perfect competition paradigma predicts. What is needed is competition and some state regulation. He makes a good case for a third way between neoliberalism and central planning.
There is no math in the book, so it can be read at many levels. It covers a broad range: Competition policy, privatization theory, forms of competion and much more. After reading it, I had a much better understanding of real world problems economies face.
On a side note: Nicholas review is simply wrong. Stiglitz employs almost only rational choice models. Problems occur because information is costly, not because people are dumb.
on March 21, 1999
Joseph Stiglitz makes a powerful argument that neither capitalism nor socialism can achieve the economically efficient outcome suggested by the neo-classical model. Neither markets nor central planners can optimally direct resources to their most productive uses. The information required to do so is simply not available.
The neo-classical model suggests that the forces of supply and demand result in equilibrium, market clearing prices - a single price for each commodity. We see all around us evidence that real world processes to not achieve this optimal end. The same items sell for different prices at different stores; it is possible for significant numbers of workers to remain unemployed for long periods of time. Stiglitz explains that this outcome reflects informational imperfections in market generated prices. It is costly for shoppers to compare prices in every store before they make purchases. Employers may pay employees higher than market clearing wages to increase worker productivity, resulting in prolonged unemployment. If market generated prices and wages were as informationally efficient as the neo-classical model suggests, Stiglitz argues that market socialism could be just as efficient as free market capitalism. Markets could be permitted to function to the degree necessary to generate prices, which central planners could use to direct the economy. Stiglitz further argues that the most critical information planners need, to plan large scale investments, are not generated by markets anyway, because the appropriate futures markets (where investors could insure against bad investments) can not exist.
Stiglitz's explanation of how the neo-classical model of constrained optimization cannot describe real world phenomena is compelling, as is his argument that both market socialism and market capitalism face problems of information and incentives. Where Stiglitz is weakest is when he casually asserts, as he often does in this book, that government intervention could resolve some of these problems under either system. He routinely asserts that government intervention could address, for instance, problems of externalities through the application of Pigouvian taxes. He does not, however, discuss how government might determine the proper tax,(in the absence of a market in the externality), or how it might insure its application in the face of special interest political pressure. In his calls for government intervention, government is treated as benevolent, omnipotent and omniscient.
Stiglitz presents a coherent argument of why market socialism failed in the real world, and further, why market capitalism, as we see it practiced around us, does not live up to the promise of the neo-classical model.
on May 31, 2001
An interesting book, perhaps worth more than a single star, but, please, how silly are people according to great economists? The premise of the market failures of the book is the inability of writing complex contracts. A smoking at home paradox disappears if you can contract with the insirance company about fires from smoking. A race-to-the-bottom collective action problem between lessors and lessees disappears if the lessee can be induced to use specific types of care--type of paint when refinishing or type of fertilizer when planting, for example.
on January 26, 1999
The author of this book has shown that market socialism would go to the dead end. China and Vietnam are not exceptional cases. Furthermore, by the approach of imperfect information, the arguments of the author against neoclassical and property theories are strongly persuadable to me.However, some political economy approach has not been mentioned in this book. I think this appproach is also useful to analyze the constraints on market socialism.