Complexity economics is a new line of economic thought that emphasizes the idea that equilibrium is not the economy’s normal state. Instead, economic agents in search of opportunities constantly adjust their actions in response to the current environment, the actions of other economic agents, and technological progress. Compared with standard economics, this approach is better able to handle interactions, change, and nonequilibrium phenomena.
What’s Inside?
Complexity economics challenges a major feature of conventional economic thinking—namely, the idea that the economy is adequately characterized by a state of equilibrium. Instead, complexity economics considers the economy to be in a constant state of motion and reconstruction. Where equilibrium economics focuses on order, determinacy, and deduction, complexity economics emphasizes contingency, indeterminacy, and openness to change. Complexity economics also requires a different methodology. The formation of outcomes and the appearance of dynamic phenomena are best studied with the help of computer-based simulations. According to the author, this shift is in line with broader scientific development, in which procedural and algorithmic approaches are emphasized over continuous and equation-based methods.
How Is This Research Useful to Practitioners?
By definition, an equilibrium approach to the economy rules out the possibilities that the system is exploited by well-positioned participants or that markets get out of control. Arguably, both elements contributed to the current financial crisis. In contrast, complexity economics teaches the importance of abrupt structural change or the actions of individual market participants. This view has important policy implications as well because supervisory control is needed to counter the inherent instability of the economic system.
The author explains the application of complexity economics to nonequilibrium phenomena with three examples. He first discusses the Santa Fe artificial stock market, a computer-based model in which computer programs develop and implement individual forecasting models and investment strategies. These strategies either succeed or die out while market prices fluctuate and spontaneous bubbles and crashes occur. His second example is a related phenomenon: clustered volatility, or generalized autoregressive conditional heteroskedasticity (GARCH) behavior. Volatility is low when agents’ forecasts are reasonably well aligned and they have little incentive to change them. Then, some agents develop superior predictors, and the overall pattern is disturbed. The ensuing period of intense readjustment causes heightened volatility.
The final example deals with contagion in the banking sector. A bank that is holding distressed assets may be forced to increase liquidity and thus to call on its counterparties. Whether this distress fades out or increases depends on the density of connections among banks. In a sparsely connected network, the effect will die out, but it can have lasting and severe effects in a densely populated network.
How Did the Author Conduct This Research?
The author discusses the relationship between complexity economics and neoclassical economics and demonstrates the main operating principles of the new approach. He draws on empirical and theoretical research published over the past 20 years. He first demonstrates that nonequilibrium is the natural state of the economy and then shows how to develop theories for this situation. The author provides several examples of dynamic phenomena under nonequilibrium and stresses the importance of positive feedback, which arises from self-reinforcing behavior. He also points to the important role of technological change, which produces novel structures.
Abstractor’s Viewpoint
The main theme in complexity economics is a focus on the unruly process of change rather than on a tidy state of equilibrium. This process of change is most fruitfully studied in the context of computer-aided simulations, which have become available only recently with advances in computing power and speed. The study of adjustment processes appears particularly relevant in cases of extreme distress, where neoclassical economics has failed to provide satisfactory explanations or warning signs.