Economics is in crisis. On one hand, behavioural economics is now well-established, but on the other hand, most economics models are still based on rational expectations with constraints, called “frictions”. The standard program adds more and more constraints to rationality in hopes that this will approximate real behaviour, but this may never work. It is increasingly clear that heterogeneity (the fact that people and institutions are diverse) is essential to understand problems such as inequality. There is a major effort to address this challenge, but the models that do this are technically complicated and rapidly become intractable as they become more realistic. Finally, there is a fundamental challenge due to the fact that we have very little historical data available to fit models for a complicated and evolving economy.
Complexity economics offers solutions to these problems. It advocates modelling behaviour in terms of heuristics and myopic reasoning, as observed in behavioural experiments. It advocates the use of simulations, making it much easier to incorporate heterogeneity in a tractable manner. Finally, it advocates using highly granular data, that accurately captures heterogeneity, to fit the models. Professor Doyne Farmer will present examples where this approach has had success, including applications to technology forecasting, economic growth and climate change, and present a vision of what it can do in the future.