Detection of imbalances, bubbles and boom-bust episodes: fundamental and non-fundamental shocks

The second work package that aims to improve financial sector modelling and eventually to deliver information on how to best incorporate such advances in policy-focused macroeconomic models. During the global financial crisis economists charged with providing model-based policy advice at European institutions had to use ad-hoc adjustments for explaining the interaction of the real economy and asset price volatility. To give an example, the link between risk premia embodied in interest rates, housing demand and house prices in Spain moved far off model-based assessments as price increases failed to dampen demand prior to the crisis.

The approach that is further developed here in order to improve the understanding of the mechanics of asset price booms and busts is related to and partly builds on the work on modelling “animal spirits” in WP2. It investigates the role of learning, bounded rationality and multiple equilibria in driving asset price volatility. Belief dynamics are modelled explicitly and empirical applications serve to evaluate the explanatory power for financial data. Explicit modelling and empirical evidence prepare the ground for including these mechanisms in policy-focused models to be used for detecting imbalances and analysing the macroeconomic consequences of fundamental and non-fundamental shocks.

Key Objectives

(O.4.1) Incorporating recent advances on explaining asset price volatility and bubble dynamics using models of learning into policy determination. How should macroeconomic policy be decided if asset prices are determined by expectations of learning?

(O.4.2) Try to elicit from the existing models how to design mechanisms that warn policy makers of possible imbalances. How to design expectation surveys and use them in order to make them useful for policy analysis.

Main participants