One of the central challenges that Darden MBA students grapple with in their first year macroeconomics course is to understand how governments can (and should) set policy to create favorable conditions for economic growth and vitality. Whether the topic is monetary vs. fiscal policy, laissez-faire vs. interventionist philosophies, or mature vs. emerging economies, a central assumption that guides Darden first year macroeconomic debate is that governments can affect economies by taking certain actions.

Paul Ormerod, in his 1998 book Butterfly Economics, preempts these debates by arguing that governments are actually unable to affect economies in any significant way, for the simple reason that in order to implement effective intervention, it is necessary for policy makers to predict what will happen with and without the actions. Ormerod argues – convincingly – that modern economic theories are unable to predict future events because 1) they do not take into account the fact that individuals within economic systems influence one another, and 2) there is far too much chaos and chance involved in the workings of national economies to make meaningful predictions.

Using the powerful metaphor of ants choosing which food source to go to, Ormerod demonstrates that individuals in a system do not simply act to maximize utility, but that they also react to the choices and actions of others. Because of the fact that at any given time individuals have various probabilities of either acting on their own or being influenced by others, it is impossible to say exactly how the system as a whole will behave in the long (or even short) term. Impossible-to-predict outcomes can arise as a result of chance events that occur very early on in the evolution of a system, and these outcomes can evolve to other scenarios with equally unpredictable speed. Moreover, where the system starts out can have a powerful effect on what will happen in response to various endogenous or exogenous (e.g., government initiated) forces – that is to say, cause-and-effect relationships are non-linear.

Ormerod also uses the ant metaphor to explain why it is impossible to predict from economic data other outcomes that depend on interactions of large groups of human beings – for example, crime or marriage rates. The unstated but obvious tie-in to an e-business course is that it is not only impossible to predict the evolution of technological platforms and environments, but also that critical points in time arise during which very small events can have large and long-lasting effects on the entire system, and that where a system starts out can be the ultimate determinant of how the market evolves.

While Ormerod is, as far as I can tell, precisely on target in his criticisms of conventional economic theory, his approach unfortunately suffers from the same problem as many of his targets: reductionism. Just as the monetarists and the Keynesians fail because their “simple” models do not take into consideration interpersonal influences, Ormerod’s model fails because he expects it to explain everything that the others leave out. In fact, common sense tells us that economies are subject to forces that go beyond simple utilities, and beyond even Ormerod’s more complex ant model. True, economists are unable to predict future events in part because of the limitations in their models that Ormerod points out. But those models – and Ormerod’s – also fall short, in my opinion, because economic outcomes are ultimately driven by individual human decisions, and human decisions are far too psychologically complex to understand within the current state of our mathematical modeling capabilities.