FOR more than 200 years since the first publications of Adam Smith, economic theorists have assumed a bounded and reasonably well-behaved world. Supply and demand regulating prices with singular long-term stability points for the majority of markets has been the overriding view. And in the colonial and developing world of the past this was more or less correct, although there have always been serious disagreements between theory and practice. One of the most obvious and notable disagreements has been the lack of coupling between the supply and demand for oil-based products. Over the past 40 years we have seen oil gluts result in high prices, and shortages predicating very low prices. But we have also had the realisation that oil supplies are finite.

Despite the various, and very obvious, anomalies, economic theory and supposed understanding are still largely dominated by linear models involving negative feedback and control. Only a few seem to have conceded that positive feedback, feed forward and multiple stability points occur, and chaos is the dominant mode. It is as if they have lived on a plain all their lives and have yet to see their first mountain. The universe, physical terrain and economics are not flat, they are lumpy and extremely varied. They are also inherently unstable in the long term.

Three principal changes make conventional economic theory dangerous for individuals, corporations and nations. First: the transition to a bit economy sees no finite limit to the creation and supply of soft products. Second: their supply and delivery can be both ubiquitous, unlimited and instantaneous. Third: soft product prices and production costs can be almost unrelated in any way. A physical product (hardware) will have a research-development-production to price ratio in the tens. In contrast, a soft product (bitware) will have such ratios in the thousands or millions. Communications, computing and the control of everything by bits instead of atoms have catapulted economics into a new domain well beyond the thinking of Adam Smith and his devotees. Economies are no longer flat, smooth, well-behaved with occasional depressions and universal stability points; they are craggy and chaotic.

As witnessed by weather prediction, modelling a chaotic world is a highly complex and difficult task that can only be contemplated by machines. To model a small cycle of a modest national economy would require the entire human race working non-stop for weeks, and never making a numerical mistake, no matter how small.

In contrast with the limited world of atoms, this new and networked world is full of opportunity, and an infinity of routes to customer and customised product. And unlike the world of the past, in this new world there is no limit to the number of bits that can be produced and communicated. It also means selling >1,000x more at <0.01x the conventional price to realise market domination, and huge incomes very quickly. But with this richness will come new surprises and new business models, because this world has a myriad of stability points, and counter-intuitive outcomes. Those who grasp this fundamental difference early will be the winners in the new millennium race.

Peter Cochrane holds the Collier Chair for the Public Understanding of Science & Technology at the University of Bristol. His home page is:
http://cochrane.org.uk