ABSTRACT
What are the advantages of regarding capitalist economies as complex, adaptive, self-organizing systems typically far from equilibrium? First, we can dispense with many of the nagging problems that arise in the
representation of firms and households in an equilibrium setting. The vision of market equilibrium requires each firm and household to be in equilibrium as a precondition of the economy to be in equilibrium. This further requires tremendous and unrealistic informational and planning capacities on the part of the firms and households. Firms and households facing risky investment opportunities have to have accurate information about prices of assets at future dates and in detailed contingencies in order to value investment plans. It has always seemed improbable that firms and households in real economies have these information-gathering and processing capacities, and a great part of the theoretical effort generated by the equilibrium research program is directed to weakening or rationalizing these requirements. Because selforganization in complex systems is robust, it requires much less structure in the behavior of firms and households to sustain it, and is compatible with a much wider range of firm and household sophistication. The innovating firms in the endogenous technical change world of Chapters 2 and 3, for example, do not have to evaluate the whole complex path of price changes their innovations set in motion, or even do much more than grope tentatively toward cost reduction. This freedom is reflected in the much more open modeling methodology
suggested by the complex systems approach. The behavior of the agents composing an economy can be both simpler and more closely adapted to particular institutional or historical situations. The researcher’s insights or
guesses as to behavior that may be critical to the evolution of the economy have wider scope in the complex systems approach. Second, the complex systems program allows us to acknowledge a much
wider range of phenomena in the study of the capitalist economy, and to situate each in a more appropriate and relevant context. For example, the various conceptions of market equilibrium, including both Classical or Neoclassical theories, assume that technology remains constant on the time-scale on which equilibrium or natural prices are formed. But we also know that innovation is central to the competitive process of capitalist economies, and that it leads to a non-stationary evolution of technology. The complex systems approach embraces this dualitywithout embarrassment. There is nothing surprising in a complex systemhaving highly organized, even quasi-equilibrium subsystems. The inherent character of complex systems is to articulate a large number of subsystems which may each be organized in quite different ways. Third, the complex systems approach should lay to rest the dubious project
of detailed forecasting of the economic future. Complex systems are computationally irreducible, or incompressible, in the sense that there is no way of encoding the evolutionary path of the system in a model less complex than the system itself. This feature of real capitalist economies presumably lies behind what appear to be a constant stream of methodological problems that confront the equilibrium research program, for example, the failure of statistical methods in the face of non-stationarity. As I have tried to argue in this book, the study of self-organization can to some degree provide a substitute for detailed forecasting, in that self-organization is responsible for the reproduction of certain key features in economic data. But the study of self-organization inherently avoids the fallacy of thinking that a model of the economy can represent its evolution in detail. Another way to put this point is that the complex systems vision restores the genuine openness of the economic future as an evolutionary process, without requiring the assumption of a definite future equilibrium path (or collection of equilibrium paths) to organize the current behavior of agents as the equilibrium program does. This may seem to be as much a disadvantage as an advantage methodologically, but if the future is genuinely open, it is better to work within a framework that acknowledges that reality. Classical political economy reflects the combination of a breathtaking
openness of vision and a formidable realism concerning human societies. The Classical political economists were fertile speculators concerning the most abstract and general features of economic life, and at the same time
insisted on following out to sometimes unpalatable conclusions the conflictual logic of capitalist social relations. We inherit from this period of intellectual creativity a powerful range of insights into class, distribution, population, and the dilemmas inherent in the long-run pursuit of capital accumulation. The neoclassical shift of emphasis to the very short-run properties of market equilibrium, and to a thermodynamic conception of equilibrium already represents a drastic curtailment of the Classical political economists’ program. The introduction of rigorous mathematical and statistical methods into the research program of economics in the twentieth century further circumscribed and narrowed the range of its discourse. Acknowledging the complexity of the economic system can perhaps restore to us the wideranging unflinchingly realistic speculative freedom the Classical political economists enjoyed.
