Complexity in economics and finance
Many economic models are based on perfectly rational agents, and don't take into account the high degree of unpredictability in human behaviour. At the IAS we aim to develop more realistic economic models to help policy-makers make better predictions.
In a perfectly rational world the economy is characterised by an average agent (consumer, producer, investor, etc.), who is a perfect optimiser with rational expectations about the future. But the real world is much more complex, and socio-economic systems are highly influenced by people’s differing expectations and adaptive behaviours.
Panics, bubbles, crashes, herding. Crises like these are nightmares for policy-makers and citizens alike. The 2008 fall of Lehman Brothers, for example, triggered a cascade of events that had a profound impact not just on banks and stock markets, but also on the incomes and well-being of people around the world.
When the crisis came, the serious limitations of existing economic and financial models immediately became apparent
- Jean-Claude Trichet, European Central Bank (2010)
Developing more realistic economic models
The goal of this research programme is to develop a more realistic financial-economic, agent-based, network model that policy-makers can use to predict or manage socio-economic crises. As a starting point, we are investigating whether complexity modelling tools that have been successful in the natural sciences can be applied to complex socio-economic systems.
Understanding the complexity of economic systems requires drawing on a wide range of disciplines including not only economics but also for example psychology, sociology, physics and computational science.
Programme director
Cars Hommes
Professor of Economic Dynamics, University of Amsterdam
My research focus is on complex systems applications in economics and finance. My work includes behavioural macroeconomics; experimental macroeconomics & finance; behavioural finance, bounded rationality; expectation formation and learning; heterogeneous agent modelling.
Research fellows
Sebastian Poledna
Plans to develop an agent-based model (ABM) for the Dutch economy that can be used for mainstream applications such as economic forecasting and macroeconomic policy analysis.
More information
To find out more about this research programme, or discuss getting involved contact the Programme Director.