While George Osborne was insisting that the UK must move from “securing financial stability” to “securing consistent growth”, another session at last week’s World Economic Forum in Davos was debating the merits (or otherwise) of current economic forecasting.

The article describes why, despite the vaguaries of the weather, economists, have a harder job than weather forecasters because we require them to see two to three years into the future rather than simply assure us that the rain will hold off for this weekend’s barbeque.

More seriously, the debate highlighted that, despite being highly complicated, current economic models based on System Dynamics (SD) are not catching all of the factors at work within the system.

SD models make predictions for the future based on past relationships and interactions. But when these relationships and interactions are between constantly changing humans, resources, war, natural disasters and technology, as they are in economics , what has happened in the past may not be an accurate predictor of the future.

All of which might explain why decision makers are still struggling to understand the feedback loops on the credit crunch.

With the chief exec of a major asset management firm estimating that his firm’s economic advisers get their predictions right “about 3 or 4 times out of ten” and Professor Robert Shiller, economics professor at Yale University stating that “The economic profession got too much in love with its models”, perhaps it’s time economists learned to love a new model.

Agent Based Modelling and Simulation delivers true understanding of the reasons behind system performance and as such can provide the insights that economists are now looking for.

In fact, last summer the Economist reported that America’s Federal Reserve and the Bank of England were exploring whether ABMS might provide a better early warning system which could prevent another economic crisis.

We’ll keep you posted as the debate, like the world economy, rumbles on.