Models, knowledge creation and their limits

Instead of modelling socio-economic situations as mechanical systems with fixed, predictable behaviour, we now see that socio-economic systems are really complex systems, in which various possible structural changes can occur giving rise to a range of different possible futures. This necessary future uncertainty automatically imposes an uncertainty on the precise pay-off that any particular action or decision that an agent may take. Because of this, the decisions that agents will make are also uncertain and this poses limits to our ability to model socio-economic systems and therefore to the knowledge that we can have at any time about the future. Because of this constant knowledge decay, what matters in real world situations of markets and business is the generation of new, current knowledge. Contrary to traditional science in which the natural laws are independent of who knows them, in social and economic systems, knowledge of system behaviour decays over time, and is in any case used up when it triggers new behaviour in the system. Several examples of evolutionary market systems are presented which demonstrate how knowledge is constantly created and destroyed, and the problem of change, innovation and design are shown to be part of a ‘boundedly rational’ view in which imperfect search gives rise to ‘good enough’ behaviour. All of this is a radical departure from the traditional approach that falsely believe in the optimisation of designs, behaviours and profits. Complexity tells us that we must accept risk and uncertainty and work loosely, keeping our options open as much as possible.